From JelyZhu  mxic.com.cn  Wed Nov  2 09:53:59 2005
From: JelyZhu  mxic.com.cn (JelyZhu@mxic.com.cn)
Date: Wed Nov  2 17:30:18 2005
Subject: [Mp4-tech] About the intra prediction mode!
Message-ID: <OF1F2F4B34.0C440030-ON482570AD.0009F549@mxic.com.cn>
Hi experts,
About the intra prediction mode I got the different algorithm from Iain
E.G.Richardson's book:"H.264 and MPEG-4 Video Compression: Video Coding for
Next-generation Multimedia".It said that "If either of these nerghbouring
blocks is not available, the corresponding value A or B is set to 2(DC
prediction mode)" and this is mismatch with JM decoder.
I took a example.
Suppose A is not available and B is available. PredmodeB is equal to 1.
as the spec, predmodeA is set equal to 2 so the mostprobableIntraPredmode
is equal to 0.
as the jm101 decoder, the mostprobableIntraPredmode will set to 2 directly.
Which one is the result? I am really confused about it.
Acctually I had take this problem last month but I still don't understand
clearly.Pls give me some hints.
Thank you in advance.
Best Regards
Jely
From agulati  NeoMagic.com  Wed Nov  2 12:27:33 2005
From: agulati  NeoMagic.com (Atul Gulati)
Date: Wed Nov  2 17:30:23 2005
Subject: [Mp4-tech][ISO-fileformat] stco box
Message-ID: <FE009gS4ycdcIeVuqvM0002d4bd@Mx1.Myoutlookonline.com>
Hi,
I am currently working on a Player for encrytped MPEG4 File. According to
the encryption standard only the samples are encrypted (actual frames) and
not meta data. But the encrpyption is done not per sample , but taking a
chunk as a block. 
>From the "stco" box i am able to get the chunk offset but not the size of
the chunk. Is there any way i can get the size of the chunk.If need the size
and the offset both for the chunks present in the MPEG4 File.
Regards,
Atul.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051102/7ac7c0a5/attachment.html
From tomar_atul2000  yahoo.com  Wed Nov  2 05:55:24 2005
From: tomar_atul2000  yahoo.com (atul tomar)
Date: Wed Nov  2 17:30:29 2005
Subject: [Mp4-tech] Doubt about coeff_token parsing with CAVLC tables..
Message-ID: <20051102135524.96089.qmail@web53905.mail.yahoo.com>
Hi Experts,
I have doubt in CAVLC decoding, specially how to decode coeff_token. Any hint or comment will be great help for me.
There are 6 VLC tables. 
We can compute nC at decoder side now how to get coff_token from bitstream then matching in particular table to know TrailingOnes and TotalCoeff.
Is there any logic to parse coeff_token like we do in exp_golomb codes.
Please give idea to parse coeff_token from bitstream.
Thanks in advance.
Regards,
Atul
-----
Hi,
I am also not able to understand the same thing.. i mean the way tables
are derived and formed... Can anyone point to some way to decode these
tables in an efficient way..... i will appreciate if someone can
provide a rough outline or code. Reference code is not using decoding
CAVLC the way it is mentioned in standard doc....although results are
same.
As for Xu Ke 's  querry...its because of the probability of occurence.
The most probable event will be allocated least bits, for high
compression efficiency. considering your example...
For example,when 0<=nC<2:
1)TrailingOnes(coeff_token)=0,TotalCoeff(coeff_token)=0,then coeff_token = "1"
   Its highly probable to have TotalCoeff = '0' so only one bit
coeff_token = '1'
2)TrailingOnes(coeff_token)=0,TotalCoeff(coeff_token)=1,then
coeff_token = "000101"
  When TotalCoeff = '1'... then the most happening is that
coefficient is '1' itself.....that means TrainlingOnes = 0 is  a
distant probability....less probable more bits.
3)TrailingOnes(coeff_token)=1,TotalCoeff(coeff_token)=1,then coeff_token = "01"
  See point 2) above
4)........
   and so on.
I think my understaning is OK.. correct if its not right.
I think this will help........
regards
Satendra
On Sat, 19 Feb 2024 19:56:06 -0800, Xu Ke <kexu ee.cuhk.edu.hk> wrote:
> Dear Experts,
> Currently I am working on CAVLC decoding.Regarding to Table9-5:coeff_token mapping to TotalCoeff(coeff_token) and TrailingOnes(coeff_token),I can't find any relations between the adjacent coeff_token.Can anybody explain to me how these coeff_token derived when the standard is being formed?
> 
> For example,when 0<=nC<2:
> 1)TrailingOnes(coeff_token)=0,TotalCoeff(coeff_token)=0,then coeff_token = "1"
> 2)TrailingOnes(coeff_token)=0,TotalCoeff(coeff_token)=1,then coeff_token = "000101"
> 3)TrailingOnes(coeff_token)=1,TotalCoeff(coeff_token)=1,then coeff_token = "01"
> 4).............
> I can't see any correlations in the above "1","000101","01".Why the coeff_token does not equal to a regular bits order such as "00""01""10" which is easier for decoding?
> 
> Many thanks in advance!
> 
> Best regards,
> Xu Ke
---------------------------------
 Yahoo! FareChase - Search multiple travel sites in one click.  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051102/63eee008/attachment.html
From hkl0220  gmail.com  Thu Nov  3 13:24:28 2005
From: hkl0220  gmail.com (Kelly Huang)
Date: Fri Nov  4 14:29:26 2005
Subject: [Mp4-tech] Question about motion vector
Message-ID: <98251ba0511031024q5227d147kfe1cd2caa0b52912@mail.gmail.com>
Dear experts,
When finding the best motion vector, we
minimize the following equation
"J(m,\lambda) = SAD(s,c(m))+\lambda R(m-p), where the rate term R(m-p) is
estimated by UVLC table.
I have searched for this table for a long long time.
I was wondering if anyone knows where could I find this
table.
Best Regards
Kelly
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051103/925daa2c/attachment.html
From ee04m115  ee.iitm.ac.in  Fri Nov  4 16:16:36 2005
From: ee04m115  ee.iitm.ac.in (ee04m115@ee.iitm.ac.in)
Date: Fri Nov  4 14:36:07 2005
Subject: [Mp4-tech] [Audio-Tech] No. of tracks in the encoded MP4 file
Message-ID: <20051104104610.M47520@mail-rly.ee.iitm.ernet.in>
Hi All,
I am using BSAC quantization mode in the MPEG4 ISO reference code. I convert
the fl4 output to mp4 format using the "convert" utility. The encoded MP4
stream contains two tracks for both mono and stereo streams. Why should it
be so?
Hope someone can throw some light on this issue.
-Deepak
--
#Open WebMail Project (http://openwebmail.org)
EE Department IIT Madras
From ee04m115  ee.iitm.ac.in  Fri Nov  4 16:17:25 2005
From: ee04m115  ee.iitm.ac.in (ee04m115@ee.iitm.ac.in)
Date: Fri Nov  4 14:36:11 2005
Subject: [Mp4-tech] [Audio-Tech] BSAC Encoder: Stereo mode issue
Message-ID: <20051104104653.M90900@mail-rly.ee.iitm.ernet.in>
Ashok,
 I too am facing problems with the stereo streams. However, if I use a 16kHz, 
8 bit stereo input, the thing works just fine. 
-Deepak
--
#Open WebMail Project (http://openwebmail.org)
EE Department IIT Madras
From singer  apple.com  Fri Nov  4 11:57:40 2005
From: singer  apple.com (Dave Singer)
Date: Sun Nov  6 14:55:31 2005
Subject: [Mp4-tech][ISO-fileformat] stco box
In-Reply-To: <FE009gS4ycdcIeVuqvM0002d4bd@Mx1.Myoutlookonline.com>
References: <FE009gS4ycdcIeVuqvM0002d4bd@Mx1.Myoutlookonline.com>
Message-ID: <p0623093abf916d94833b@[17.202.35.52]>
At 12:27  +0530 2/11/05, Atul Gulati wrote:
>Hi,
>
>I am currently working on a Player for encrytped MPEG4 File. 
>According to the encryption standard only the samples are encrypted 
>(actual frames) and not meta data. But the encrpyption is done not 
>per sample , but taking a chunk as a block.
>
>From the "stco" box i am able to get the chunk offset but not the 
>size of the chunk. Is there any way i can get the size of the 
>chunk.If need the size and the offset both for the chunks present in 
>the MPEG4 File.
The sampel to chunk box tells you how many samples are in each chunk, 
and the sample size box tells you how big each is;  sum the sizes of 
the samples...
>
>Regards,
>Atul.
>
>_______________________________________________
>NOTE: Please use clear subject lines for your posts. Include [audio, 
>[video], [systems], [general] or another apppropriate identifier to 
>indicate the type of question you have.
>
>Note: Conduct on the mailing list is subject to the Antitrust 
>guidelines found at 
>http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-- 
David Singer
Apple Computer/QuickTime
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051104/cbd107cf/attachment.html
From zombiyaga  yahoo.com  Sat Nov  5 13:27:14 2005
From: zombiyaga  yahoo.com (Alex)
Date: Sun Nov  6 14:55:36 2005
Subject: [Mp4-tech][Video] Short video header
In-Reply-To: <4112AEBC.D9474322@iis.fhg.de>
Message-ID: <20051105212714.97111.qmail@web53910.mail.yahoo.com>
Hi All, 
Do you have any samples of MPEG-4 (short header mode)
? 
thanks, 
--- Herbert Thoma <tma@iis.fhg.de> wrote:
> Sandip Parikh wrote:
> > 
> > Hi,
> > 
> > Could someone please clarify short_video_header
> for me:
> > 
> > At the elementary stream level
> (video_plane_with_short_header() and
> > below) is a MPEG-4 stream with short_video_header
> = 1 identical to a
> > H.263 stream in syntax or there is a difference ?
> 
> MPEG-4 short_video_header is identical to h.263
> baseline, i.e. h.263
> without any annex.
> 
> Regards,
>  Herbert.
> 
> > Thank you,
> > 
> > Regards,
> > 
> > Sandip Parikh
> > 
> > _______________________________________________
> > NOTE: Please use clear subject lines for your
> posts. Include [audio, [video], [systems], [general]
> or another apppropriate identifier to indicate the
> type of question you have.
> > 
> > Note: Conduct on the mailing list is subject to
> the Antitrust guidelines found at
>
http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
> _______________________________________________
> NOTE: Please use clear subject lines for your posts.
> Include [audio, [video], [systems], [general] or
> another apppropriate identifier to indicate the type
> of question you have.
> 
> Note: Conduct on the mailing list is subject to the
> Antitrust guidelines found at
>
http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
> 
--
Regards, 
Alex
__________________________________ 
Yahoo! FareChase: Search multiple travel sites in one click.
http://farechase.yahoo.com
From pankaj_bajpai_iet  operamail.com  Tue Nov  8 08:44:05 2005
From: pankaj_bajpai_iet  operamail.com (pankaj bajpai)
Date: Tue Nov  8 16:02:05 2005
Subject: [Mp4-tech][Video] Short video header
Message-ID: <20051108074405.422A17B4E4@ws5-10.us4.outblaze.com>
Hi Alex!!
Some of the conformance bitstrems for Mpeg4 Shortvideoheader are available at 
http://standards.iso.org/ittf/PubliclyAvailableStandards/ISO_IEC_14496-4_2004_Conformance_Testing/video_conformance/version_1/simple/short.zip
In this zip floder, following 5 sequences I think have some errors as the Microsoft decoder (July 3rd 2000 version 2 FDAM1-2.3-001213) fails on the following streams:
1. short\hit035.m4v
2. short\mit021.m4v
3. short\mit022.m4v
4. short\mit023.m4v
5. short\mit024.m4v
I guess the error is that the resync marker are appearing at not align boundary.
If some one has more bitstreams, then please share them.
With regards
pankaj
From: Alex <zombiyaga@yahoo.com>
To: tma@iis.fhg.de Sandip Parikh <sparikh@sarnoff.com>
Cc: mp4-tech@lists.mpegif.org
Subject: Re: [Mp4-tech][Video] Short video header
Date: Sat, 5 Nov 2023 13:27:14 -0800 (PST)
Hi All,
Do you have any samples of MPEG-4 (short header mode)
?
thanks,
----- Original Message -----
From: mp4-tech-request@lists.mpegif.org
To: mp4-tech@lists.mpegif.org
Subject: Mp4-tech Digest, Vol 28, Issue 3
Date: Sun, 6 Nov 2023 12:13:02 -0500 (EST)
> 
> Send Mp4-tech mailing list submissions to
> 	mp4-tech@lists.mpegif.org
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	http://lists.mpegif.org/mailman/listinfo/mp4-tech
> or, via email, send a message with subject or body 'help' to
> 	mp4-tech-request@lists.mpegif.org
> 
> You can reach the person managing the list at
> 	mp4-tech-owner@lists.mpegif.org
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Mp4-tech digest..."
>
> 
> Today's Topics:
> 
>     1. Re: [Mp4-tech][ISO-fileformat] stco box (Dave Singer)
>     2. Re: [Mp4-tech][Video] Short video header (Alex)
From: Dave Singer <singer@apple.com>
To: agulati@neomagic.com mp4-tech@lists.mpegif.org
Subject: Re: [Mp4-tech][ISO-fileformat] stco box
Date: Fri, 4 Nov 2023 11:57:40 -0800
>
> 
> At 12:27  +0530 2/11/05, Atul Gulati wrote:
> > Hi,
> >
> > I am currently working on a Player for encrytped MPEG4 File. 
> > According to the encryption standard only the samples are 
> > encrypted (actual frames) and not meta data. But the encrpyption 
> > is done not per sample , but taking a chunk as a block.
> >
> > From the "stco" box i am able to get the chunk offset but not the 
> > size of the chunk. Is there any way i can get the size of the 
> > chunk.If need the size and the offset both for the chunks present 
> > in the MPEG4 File.
> 
> The sampel to chunk box tells you how many samples are in each 
> chunk, and the sample size box tells you how big each is;  sum the 
> sizes of the samples...
> 
> >
> > Regards,
> > Atul.
> >
> > _______________________________________________
> > NOTE: Please use clear subject lines for your posts. Include 
> > [audio, [video], [systems], [general] or another apppropriate 
> > identifier to indicate the type of question you have.
> >
> > Note: Conduct on the mailing list is subject to the Antitrust 
> > guidelines found at 
> > http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
> 
> 
> -- David Singer
> Apple Computer/QuickTime
From: Alex <zombiyaga@yahoo.com>
To: tma@iis.fhg.de Sandip Parikh <sparikh@sarnoff.com>
Cc: mp4-tech@lists.mpegif.org
Subject: Re: [Mp4-tech][Video] Short video header
Date: Sat, 5 Nov 2023 13:27:14 -0800 (PST)
>
> 
> Hi All,
> 
> Do you have any samples of MPEG-4 (short header mode)
> ?
> thanks,
> 
> 
> --- Herbert Thoma <tma@iis.fhg.de> wrote:
> 
> > Sandip Parikh wrote:
> > > > Hi,
> > > > Could someone please clarify short_video_header
> > for me:
> > > > At the elementary stream level
> > (video_plane_with_short_header() and
> > > below) is a MPEG-4 stream with short_video_header
> > = 1 identical to a
> > > H.263 stream in syntax or there is a difference ?
> >
> > MPEG-4 short_video_header is identical to h.263
> > baseline, i.e. h.263
> > without any annex.
> >
> > Regards,
> >  Herbert.
> >
> > > Thank you,
> > > > Regards,
> > > > Sandip Parikh
> > > > _______________________________________________
> > > NOTE: Please use clear subject lines for your
> > posts. Include [audio, [video], [systems], [general]
> > or another apppropriate identifier to indicate the
> > type of question you have.
> > > > Note: Conduct on the mailing list is subject to
> > the Antitrust guidelines found at
> >
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
> > _______________________________________________
> > NOTE: Please use clear subject lines for your posts.
> > Include [audio, [video], [systems], [general] or
> > another apppropriate identifier to indicate the type
> > of question you have.
> >
> > Note: Conduct on the mailing list is subject to the
> > Antitrust guidelines found at
> >
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
> >
> 
> 
> --
> Regards,
> Alex
> 
> 
> 		
> __________________________________
> Yahoo! FareChase: Search multiple travel sites in one click.
> http://farechase.yahoo.com
>
> 
> _______________________________________________
> Please use clear subject lines for your posts. Include [audio, 
> [video], [systems], [general] or another apppropriate identifier to 
> indicate the type of question you have.
> 
> Conduct on the mailing list is subject to the Antitrust guidelines 
> found at 
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-- 
_______________________________________________
Surf the Web in a faster, safer and easier way:
Download Opera 8 at http://www.opera.com
Powered by Outblaze
From gvgrao_spl  yahoo.com  Tue Nov  8 01:07:16 2005
From: gvgrao_spl  yahoo.com (venu rao)
Date: Tue Nov  8 16:02:10 2005
Subject: [Mp4-tech] Regarding conformance testing of HE-AAC
Message-ID: <20051108090716.57316.qmail@web53206.mail.yahoo.com>
Hi all,
I am presently working on MPEG-4 HE-AAC.
I am having some problem in the conformance testing of _twi vectors.
The vector showing that it's of AAC MAIN profile.
My decoder is not decoding that particular vector(as it supports only LC profile).
Whether my decoder should pass conformance test for _twi vectors. As no particular criteria is given in conformane document.
If any one knows about this issue, please mail me.
Thanks in advance,
Venu
__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051108/a4efd3a0/attachment.html
From golnaz.ab  gmail.com  Tue Nov  8 07:32:53 2005
From: golnaz.ab  gmail.com (golnaz Abdollahian)
Date: Tue Nov  8 16:02:15 2005
Subject: [Mp4-tech] mb_type
Message-ID: <1d98a6bd0511080732q7c98c948w60fa5f1fccd4826f@mail.gmail.com>
Hello,
I am wondering if you can help me with this problem:
Can i define mb_type from some input file as additional input to encoder?
I want to label some of macro blocks in some frames as B_Skip or P_Skip. Is
there a way to do it by setting the configuration file and give some
additional information about the location of those MB 's?
I really appreciate your help,
Mina
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051108/dbcf1681/attachment.html
From tummalakalyani  gmail.com  Tue Nov  8 22:15:04 2005
From: tummalakalyani  gmail.com (tummala kalyani)
Date: Wed Nov  9 14:18:50 2005
Subject: [Mp4-tech] GOP concept in H.264
Message-ID: <13936ad50511080845r705dfeb4peb4ea91e3b89da0c@mail.gmail.com>
Hi All,
I would like to know about GOP concept in H.264. In JM encoder config file
there is a parameter "support for open GOP". How can I use this? I tried to
1-enable and 0-disable options. But I don't find any difference.
 There is provision to set I frame distance and number of B frames. When I
treid to set number of B frames more than 0 , it gives me an error message
saying that "Number of B frames should me more than frames skipped in the
input". What is the use of Frames skipped parameter in the encoder.cfg file
in JM?
 I need bitrate vs psnr comparison graphs for H.264 codec and MPEG-2 codec.
For that I am running JM and MPEG-2 reference softwares with foreman.yuv as
input video. I found that for baseline and main profiles H.264 codec output
file size is double to MPEG-2 file size. I use this filesize to calculate
bitrate( filesize * (frame rate / total frames)). But according to specs.
H.264 codec bitrate should be half to that of MPEG-2. But I am getting in
reverse. I am confused by this. I think I have not understood it properly.
Can some one explain me this clearly?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051108/7063c9f2/attachment.html
From Andreas.Schneider  codingtechnologies.com  Wed Nov  9 01:43:21 2005
From: Andreas.Schneider  codingtechnologies.com (Andreas Schneider)
Date: Wed Nov  9 14:27:29 2005
Subject: [Mp4-tech] Regarding conformance testing of HE-AAC
In-Reply-To: <20051108090716.57316.qmail@web53206.mail.yahoo.com>
Message-ID: <OF81ECB9F8.F2F9D761-ONC12570B4.0001D470-412570B4.0003F880@codingtechnologies.com>
Hello,
I don't know about the source of your conformance bitstreams, but the ones 
on the official site 
ftp://mpaudconf:adif2mp4@ftp.iis.fhg.de/mpeg4audio-conformance/compressedMp4/ 
have an AudioObjectType of 2, which is AAC_LC, so they should be fine.
Whether or not you need to test these sequences is an entirely different 
matter. The conformance document (ISO/IEC 14496-4:2004 Amd8) says:
"
If a complex QMF filter bank with a modified internal phase angle 
(hereinafter referred to as twiddles) is used in the decoder under test, 
the al_sbr_twi_* sequences shall be tested first. [...]"
"
In english: If your QMF implementation has the same phase-properties as 
the one in the reference software, you don't need to test with these 
bitstreams.
Best regards,
Andreas
mp4-tech-bounces@lists.mpegif.org wrote on 08/11/2023 10:07:16:
> Hi all,
> 
> I am presently working on MPEG-4 HE-AAC.
> I am having some problem in the conformance testing of _twi vectors.
> The vector showing that it's of AAC MAIN profile.
> My decoder is not decoding that particular vector(as it supports 
> only LC profile).
> 
> Whether my decoder should pass conformance test for _twi vectors. As
> no particular criteria is given in conformane document.
> 
> If any one knows about this issue, please mail me.
> 
> Thanks in advance,
> Venu
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around 
> http://mail.yahoo.com _______________________________________________
> NOTE: Please use clear subject lines for your posts. Include [audio,
> [video], [systems], [general] or another apppropriate identifier to 
> indicate the type of question you have.
> 
> Note: Conduct on the mailing list is subject to the Antitrust 
> guidelines found at http://www.mpegif.org/public/documents/vault/mp-
> out-30042-Antitrust.php
--
Andreas Schneider, Research Engineer
Coding Technologies GmbH
Deutschherrnstr. 15-19
90429 Nuernberg, Germany
phone: +49 (0) 911 92891 -26
fax: +49 (0) 911 92891 -99
mailto:snd@CodingTechnologies.com
From sven-  gmx.de  Wed Nov  9 13:33:58 2005
From: sven-  gmx.de (Sven Wontroba)
Date: Wed Nov  9 14:31:31 2005
Subject: [Mp4-tech] advance multimedia with mpeg-4
Message-ID: <1559219152.20051109133358@gmx.de>
Hello experts
For a while now I am researching the area of multimedia in mobile
devices and set top boxed for my end project at my university. So I
found different IST - EU research project using advanced MPEG-4
functionality like BIFS, advanced video coding for scaleability,...
Something I'm wondering about it, I see during the last years an
upcoming trend in the use of MPEG-4 video, but in real world product
(not research project) just the use of "basic" and in some new products
avc video. I know there is the GPAC project which also seems to be
used in many research projects. But I haven't seen any commercial
product out there using this advanced features in any device.
One point I see so far is, up till now you have to create xlets and
midlets and program the multimedia part if you want to be able to play
multimedia content with the existing products and create you own
parser. I don't know you can just download a module which is able to
interpret BIFS,... information over dvb or Internet? Or would it be a
program like a player you have to install on the machine in advance or
is installed already on the STB, mobile?
I haven't seen them in real world product. Or I might not be aware of
the product. So if you are aware of them please send me some
information. Additional comments are also appreciated.
I'll send this to different lists, hopefully to get some comments so I
apologize if some of you will receive this 2 times.
Thanks in advance Sven
From igino.manfre  tiscali.it  Wed Nov  9 16:51:10 2005
From: igino.manfre  tiscali.it (igino manfre')
Date: Thu Nov 10 15:41:28 2005
Subject: R: [Mp4-tech] advance multimedia with mpeg-4 - The great risk ...
In-Reply-To: <1559219152.20051109133358@gmx.de>
Message-ID: <NKENIOLHIJGIBFPJNIAIOEKHCGAA.igino.manfre@tiscali.it>
Hi Sven,
I suffer of a nightmare that -- for many aspects -- may come from questions
similar to yours.
Today broadcasting is going toward MPEG-4 (better H264, but wasn't it also
called "part 10"?): SetTopBoxes to dress the DVB logo are requested to be
high profile compliant if HighDefinition and main profile compliant if
StandardDefinition video.
The main manufacturer of compression hardware have abandoned the MPEG-2
production preferring to invest their money in H264.
Ok a single square object is sufficient, simpler and maybe better. But we
all know that H264 is not alone. "Something" surround it. We all know that
the remaining 90% of that damned 14496 standard (killed in 2002 by MPEG-LA
with that terrible licensing scheme, almost aborted) is related to
interactivity. It could linearly applied to broadcasting but it seem not to
be...
We're all are going to see (maybe to buy) SetTopBoxes able to reproduce H264
riding DVB-S2 transport sided by MHP.
Yes all the work on objects and scenes seems to be ignored.
What MHP does in this case ? Not known. A generic interest, but since
interactivity may come after (like love in he marriages), the systems on
chips expected for december will include "only" h264 Frext player, but no
word about management of MPEG-4 and xmt wrappers.
STBs manufacturers deal with consumer products. They push what the market
analysers forecast for the near future (hopefully for today). DVB and MHP
could do a lot of things about, but they seems not aware of this.
This -- unlucky -- is not only a nightmare: it is any day passing more and
more what we are going to implement.
Bye,
Igino
-----------------------
Igino Manfre'  - igino.manfre@tiscali.it <mailto:igino.manfre@tiscali.it>
Broadcast Video Consultant
Mobile (+39) 3358235346
Home   (+39) 0650916416
13, Largo Nearco
I 00124 ROMA - ITALY
-----Messaggio originale-----
Da: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org]Per conto di Sven Wontroba
Inviato: mercoledi 9 novembre 2005 13:34
A: mp4-tech@lists.mpegif.org; mpeg4@lists.infotoday.com
Oggetto: [Mp4-tech] advance multimedia with mpeg-4
Hello experts
For a while now I am researching the area of multimedia in mobile
devices and set top boxed for my end project at my university. So I
found different IST - EU research project using advanced MPEG-4
functionality like BIFS, advanced video coding for scaleability,...
Something I'm wondering about it, I see during the last years an
upcoming trend in the use of MPEG-4 video, but in real world product
(not research project) just the use of "basic" and in some new products
avc video. I know there is the GPAC project which also seems to be
used in many research projects. But I haven't seen any commercial
product out there using this advanced features in any device.
One point I see so far is, up till now you have to create xlets and
midlets and program the multimedia part if you want to be able to play
multimedia content with the existing products and create you own
parser. I don't know you can just download a module which is able to
interpret BIFS,... information over dvb or Internet? Or would it be a
program like a player you have to install on the machine in advance or
is installed already on the STB, mobile?
I haven't seen them in real world product. Or I might not be aware of
the product. So if you are aware of them please send me some
information. Additional comments are also appreciated.
I'll send this to different lists, hopefully to get some comments so I
apologize if some of you will receive this 2 times.
Thanks in advance Sven
_______________________________________________
NOTE: Please use clear subject lines for your posts. Include [audio,
[video], [systems], [general] or another apppropriate identifier to indicate
the type of question you have.
Note: Conduct on the mailing list is subject to the Antitrust guidelines
found at
http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
From alexis.tourapis  dolby.com  Wed Nov  9 09:55:11 2005
From: alexis.tourapis  dolby.com (Tourapis, Alexis)
Date: Thu Nov 10 15:41:33 2005
Subject: [Mp4-tech] GOP concept in H.264
Message-ID: <7272EE229DA1AA48B47EBDC47EB0C68B019D9157@sapphire.dolby.net>
Dear Tummala,
You may wish to refer to JVT-Q042 with regards of how to use the JM reference software. You may download this from http://ftp3.itu.int/av-arch/jvt-site/2005_10_Nice/JVT-Q042-L.zip
The parameter "EnableOpenGOP" essentially tries to emulate GOP structures (Open vs Closed GOP) used in MPEG-2 and older standards. I.e. If you wish to encode a sequence while having a 15 frame intra period, this parameter enforces a special encoding mode in the JM where inter predicted slices after the intra coded frame are not allowed to consider as references frames prior to that intra coded frame. However, this excludes non reference frames prior to that frame (in MPEG-2 this definition was explicitly used for B frames).
NumberBFrames has to be used in combination with FrameSkip. i.e. if you wish to enter two B coded frames, then you need to set both parameters to 2. Please read JVTQ042 for more info. FrameSkip may sound a bit confusing, but this was the way this parameter was used from the early stages of the JM and therefore we did not try to alter its behavior.
You will definitely find more parameters that could be of use to you in JVT-Q042, including how to set up hierarchical encoding.
I can't comment on your results since I do not know how you are doing your simulations. Obviously at this point you are not using B slices, maybe you are enforcing too many intra frames, RDO usage etc. In any case, hopefully all your questions would be answered after reviewing that document.
Best regards,
Alexis
  _____  
From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of tummala kalyani
Sent: Tuesday, November 08, 2023 8:45 AM
To: mp4-tech@lists.mpegif.org
Subject: [Mp4-tech] GOP concept in H.264
Hi All,
I would like to know about GOP concept in H.264. In JM encoder config file there is a parameter "support for open GOP". How can I use this? I tried to 1-enable and 0-disable options. But I don't find any difference. 
There is provision to set I frame distance and number of B frames. When I treid to set number of B frames more than 0 , it gives me an error message saying that  "Number of B frames should me more than frames skipped in the input". What is the use of Frames skipped parameter in the encoder.cfg file in JM?
I need bitrate vs psnr comparison graphs for H.264 codec and MPEG-2 codec. For that I am running JM and MPEG-2 reference softwares with foreman.yuv as input video. I found that for baseline and main profiles H.264 codec output file size  is double to MPEG-2 file size. I use this filesize to calculate bitrate( filesize * (frame rate / total frames)). But according to specs. H.264 codec bitrate should be half to that of MPEG-2. But I am getting in reverse. I am confused by this. I think I have not understood it properly. Can some one explain me this clearly?
-----------------------------------------
This message (including any attachments) may contain confidential
information intended for a specific individual and purpose.  If you are
not the intended recipient, delete this message.  If you are not the
intended recipient, disclosing, copying, distributing, or taking any
action based on this message is strictly prohibited.
From garysull  windows.microsoft.com  Wed Nov  9 10:53:47 2005
From: garysull  windows.microsoft.com (Gary Sullivan)
Date: Thu Nov 10 15:41:39 2005
Subject: [Mp4-tech] GOP concept in H.264
Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2461170F8F9@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
I suggest reading the reference software manual found as document
JVT-Q042 in the 2005_10_Nice directory on the JVT document site
http://ftp3.itu.int/av-arch/jvt-site.
The standard itself does not actually include a definition of a concept
called a GOP.  When people refer to "open GOP", they are usually talking
about non-reference pictures that occur in the bitstream after an Intra
picture that are displayed prior to the Intra picture being allowed to
use pictures that precede the Intra picture in decoding order as
references for their motion compensation prediction decoding process.
You sound a bit confused about what the standard does, or perhaps it is
just a question of phrasing.  I must warn you that the reference
software is really meant for those who have a significant amount of
familiarity with video compression technology.  Regarding your confusion
over why the JM reference software might use more bits to code a video
clip than some MPEG-2 reference software, could it be that the two
encoded video clips do not have comparable decoded picture quality?
Targeting equal output video quality is a fundamental part of claims of
relative bit rate savings performance capability.
Best Regards,
Gary Sullivan
________________________________
	From: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of tummala kalyani
	Sent: Tuesday, November 08, 2023 8:45 AM
	To: mp4-tech@lists.mpegif.org
	Subject: [Mp4-tech] GOP concept in H.264
	Hi All,
	I would like to know about GOP concept in H.264. In JM encoder
config file there is a parameter "support for open GOP". How can I use
this? I tried to 1-enable and 0-disable options. But I don't find any
difference. 
	There is provision to set I frame distance and number of B
frames. When I treid to set number of B frames more than 0 , it gives me
an error message saying that  "Number of B frames should me more than
frames skipped in the input". What is the use of Frames skipped
parameter in the encoder.cfg file in JM?
	I need bitrate vs psnr comparison graphs for H.264 codec and
MPEG-2 codec. For that I am running JM and MPEG-2 reference softwares
with foreman.yuv as input video. I found that for baseline and main
profiles H.264 codec output file size  is double to MPEG-2 file size. I
use this filesize to calculate bitrate( filesize * (frame rate / total
frames)). But according to specs. H.264 codec bitrate should be half to
that of MPEG-2. But I am getting in reverse. I am confused by this. I
think I have not understood it properly. Can some one explain me this
clearly?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051109/f89a301e/attachment.html
From bheemarjun  gmail.com  Thu Nov 10 11:08:06 2005
From: bheemarjun  gmail.com (arjun)
Date: Thu Nov 10 15:41:43 2005
Subject: [Mp4-tech] Measuring Video Distortion
Message-ID: <3fadd97c0511092138w22fe9c46kf487b151570fe91d@mail.gmail.com>
Dear mailing-list members,
I am working on Video transport in Ad hoc wireless n/ws. Due to
time-varying network topology (due to mobility of nodes) I am using
Multiple Description Coding (MDC) and sending MDC descriptions along
multiple paths to the destination node. At the destination I wanna
measure Distortion (e.g., Mean Square Error) for the folowing cases:
1. What is the distortion when the destination node receives i out of
m descriptions transmitted by the source?
2. What are  the network parameters that affect distortion?
Kindly provide me with some pointers.
Thanks inadvance.
Sincerely
Arjun Reddy
--
T. Bheemarjuna Reddy
Ph.D. Student
HPCN Lab
Dept. of Computer Science & Engg.
IIT Madras, INDIA
Mobile: +91-9444206507
http://hpcn.cs.iitm.ernet.in
From humaira.nisar  gmail.com  Thu Nov 10 16:54:11 2005
From: humaira.nisar  gmail.com (Humaira)
Date: Thu Nov 10 15:41:48 2005
Subject: [Mp4-tech] Iprof --Complexity analysis
Message-ID: <139c4c630511092354p28762ecer89bf38e786213ec0@mail.gmail.com>
Hi
 Do anyone has an idea how to get iprof. Or is there some other good
software that helps in the complexity analysis of motion estimation
algorithms.
Also i would like to know about some place that will help in the programming
of motion estimation algorithms
 Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051110/e0d6f897/attachment.html
From sven-  gmx.de  Thu Nov 10 10:53:28 2005
From: sven-  gmx.de (Sven Wontroba)
Date: Thu Nov 10 15:41:53 2005
Subject: [Mp4-tech] advance multimedia with mpeg-4 - The great risk ...
In-Reply-To: <NKENIOLHIJGIBFPJNIAIOEKHCGAA.igino.manfre@tiscali.it>
References: <1559219152.20051109133358@gmx.de>
	<NKENIOLHIJGIBFPJNIAIOEKHCGAA.igino.manfre@tiscali.it>
Message-ID: <433909065.20051110105328@gmx.de>
Hallo
Thanks for your comments.
> I suffer of a nightmare that -- for many aspects -- may come from
> questions similar to yours.
I'm sure there are also others asking more or less this questions, but
so far on this lists I haven't seen to much information about
multimedia and lets call it advance feature implementations. May be
there is a better placer to do so???
I guess lots of people try to implement the video part and I'm sure
it's a challenge and the demand for the advanced features? I don't
have any statistics I personally, from a authoring and multimedia
creating perspective would like to see more action in this area. On
the other side the advanced multimedia possibilitys within streaming
(smile,... ) are not used much till now. Seems like it's changing a
bit due to cheap dsl flatrates,..  still the multimedia format is
handled differently inside the players.(ms, real, quicktime,
ambulant,...) ;)ok back to mpeg4,...
I think envivo and also other doing interesting things and hope to see
more in this area.
> The main manufacturer of compression hardware have abandoned the MPEG-2
> production preferring to invest their money in H264.
seem like, well and with the two existing formats <ironic> it going to
be more fun</ironic> tandberg for example,.. seem to concentrate on
both.
> Ok a single square object is sufficient, simpler and maybe better. But we
> all know that H264 is not alone. "Something" surround it. We all know that
> the remaining 90% of that damned 14496 standard (killed in 2002 by MPEG-LA
> with that terrible licensing scheme, almost aborted) is related to
> interactivity. It could linearly applied to broadcasting but it seem not to
> be...
ok I haven't read anything about licensing the rest of the 90% but if
I look at research projects they are using it. Additionally TV-A and
MPEG-7/21 the combination and integration of mpeg-4 in there is an
other problem I see.
Also the development like LASeR is quite nice (more the mobile area).
Well if you look at the content creating side its again different. Ok
if you want to accomplish something different you have to create
something different.
>From an multimedia authoring perspective (I know you can visualize
thing in a different way as you save them) there are lots of open
questions. (or is it just me) I haven't seen so far information about
licensing the newer standards?
> We're all are going to see (maybe to buy) SetTopBoxes able to reproduce H264
> riding DVB-S2 transport sided by MHP.
the government support in the south was good to spread SetTopBoxes in
the first place. Updating to the next version might no be included.
Also the change to HDTV will be interesting even I was reading that
just payTV might change to the MPEG-4 version of it.
> This -- unlucky -- is not only a nightmare: it is any day passing more and
> more what we are going to implement.
Well I'm really impressed, about what's possible with MHP, and the
implementation of more functionality in mpeg-4 would be somehow a
little contrariar to it. Even if I think they could support each other
some features might be interesting inside the content. Some are part
of the interface adapted regarding the device it is used on and
interfaces for additional services.
I was and am still interested about a professional comment regarding
the mpeg-4 multimedia up to 21 integration in this area. I'm sure this
area will be interesting for the next ages.
> Bye,
> Igino
Thanks again for you comments
Sven
From dsn2603  rediffmail.com  Thu Nov 10 12:06:28 2005
From: dsn2603  rediffmail.com (sakthi  narayanan)
Date: Thu Nov 10 15:41:58 2005
Subject: [Mp4-tech] Regarding conformance testing of HE-AAC
Message-ID: <20051110120613.5865.qmail@webmail47.rediffmail.com>
Hi,
  Presently,Iam working in MPEG4 AAC-LC Decoder.I had the standard bitstreams.Iam having small doubt in the profile.
In Adts file,we will get the profile information from the header itself.So,I come to know the bitstream is whether LC,Main or SSR.
But,for raw how can we know whether the bitstream is LC,Main or SSR.
Whether profile information is needed for the raw file format.So,how can to differentiate the LC & main profile bitstreams in MPEG4 AAC for
raw file format.
 And also,where can i get Erroneous bitsream & required documents for the error conformace test.
 Awaiting for ur reply.
With Regards,
Sakthi Narayanan.D
On Wed, 09 Nov 2023 Andreas Schneider wrote :
>Hello,
>
>I don't know about the source of your conformance bitstreams, but the ones
>on the official site
>ftp://mpaudconf:adif2mp4@ftp.iis.fhg.de/mpeg4audio-conformance/compressedMp4/
>have an AudioObjectType of 2, which is AAC_LC, so they should be fine.
>
>Whether or not you need to test these sequences is an entirely different
>matter. The conformance document (ISO/IEC 14496-4:2004 Amd8) says:
>"
>If a complex QMF filter bank with a modified internal phase angle
>(hereinafter referred to as twiddles) is used in the decoder under test,
>the al_sbr_twi_* sequences shall be tested first. [...]"
>"
>In english: If your QMF implementation has the same phase-properties as
>the one in the reference software, you don't need to test with these
>bitstreams.
>
>Best regards,
>
>Andreas
>
>mp4-tech-bounces@lists.mpegif.org wrote on 08/11/2023 10:07:16:
>
> > Hi all,
> >
> > I am presently working on MPEG-4 HE-AAC.
> > I am having some problem in the conformance testing of _twi vectors.
> > The vector showing that it's of AAC MAIN profile.
> > My decoder is not decoding that particular vector(as it supports
> > only LC profile).
> >
> > Whether my decoder should pass conformance test for _twi vectors. As
> > no particular criteria is given in conformane document.
> >
> > If any one knows about this issue, please mail me.
> >
> > Thanks in advance,
> > Venu
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam? Yahoo! Mail has the best spam protection around
> > http://mail.yahoo.com _______________________________________________
> > NOTE: Please use clear subject lines for your posts. Include [audio,
> > [video], [systems], [general] or another apppropriate identifier to
> > indicate the type of question you have.
> >
> > Note: Conduct on the mailing list is subject to the Antitrust
> > guidelines found at http://www.mpegif.org/public/documents/vault/mp-
> > out-30042-Antitrust.php
>
>--
>Andreas Schneider, Research Engineer
>
>Coding Technologies GmbH
>Deutschherrnstr. 15-19
>90429 Nuernberg, Germany
>phone: +49 (0) 911 92891 -26
>fax: +49 (0) 911 92891 -99
>
>mailto:snd@CodingTechnologies.com
>
>_______________________________________________
>NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have.
>
>Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051110/95e15d58/attachment.html
From dmitriy  graphics.cs.msu.ru  Fri Nov 11 01:16:08 2005
From: dmitriy  graphics.cs.msu.ru (Dmitriy Vatolin)
Date: Fri Nov 11 16:00:34 2005
Subject: [Mp4-tech] List of MPEG-4 AVC/H.264 video codecs
Message-ID: <165720769.20051111011608@graphics.cs.msu.ru>
Hello!
   See below updated list of 61 companies and groups working on
   H.264/AVC codecs.
   Updates (for 2 month):
   + KDDI R&D Labs
   + Hughes Network Systems, LLC
   + Conexant Systems, Inc.
   + Sigma Designs
   + Cradle Technologies, Inc.
   + STMicroelectronics
   + PacketVideo Corp.
   + SONY Video Editor Vegas 6c with H.264/AVC support
   + PowerEncoder MPEG-4 AVC
   + Imagicity optimised H.264/AVC codec
   + Image Converter 2 Plus
   + some fixes
   King thanks for your new info messages to me!
   Site expanded copy of this list available from
   http://www.compression-links.info/MPEG-4_AVC_H264
============================================================================
LIST of MPEG-4 AVC/H.264 VIDEO CODECS
(Last updated 10 November 2023 - 61 items now!)
============================================================================
    ----------------------------------------------------------
    SOFTWARE CODECS:
    ---------------------
   1. Ateme H.264/MPEG-4 AVC Codec (Ateme)
      No download on original site:
      http://www.ateme.com/products/h264.php
      See 30 day trial of Nero Digital (with Nero Recode) available from:
      http://www.nerodigital.com/eng/index.html
      ---
   2. VSS H.264 Video Codec (Vanguard Software Solutions, Inc.)
      Trial available, VfW.
      http://www.vsofts.com/h264/codecs.html
      ---
   3. Mainconcept H.264 (MainConcept AG)
      Trial available on download page
      http://www.mainconcept.com/h264_encoder.shtml
      Mainconcept declare partnership with Elecard in H.264 in March 2005.
      ---
   4. Elecard OneClick Compressor & Elecard AVC/H.264 Decoder Package (Elecard)
      21-day free trial. Have H.264 SDK
      http://www.elecard.com/products/oneclick.shtml
      ---
   5. Moonlight H.264 Video Codec (Moonlight Cordless LTD.)
      21-day evaluation, DirectX filter.
      http://www.moonlight.co.il/products/consumer/oneclickavc/
      This codec was developed with Elecard team.
      ---
   6. SONY Video Editor Vegas 6c with AVC support (Sony Corp.) NEW!!!
      30-days evaluation.
      http://www.sonymediasoftware.com/products/showproduct.asp?PID=965
      ---
   7. QuickTime 7 H.264 (Apple Computer, Inc.)
      Integrated. Who test it? Any remarks welcome!
      http://www.apple.com/quicktime/technologies/h264/
      ---
   8. FastVDO H.264 (FastVDO LLC)
      Demo has decode time limit of 5 min.
      http://www.fastvdo.com/H.264.html
      ---
   9. LEAD H.264 Video codec (LEAD Technologies)
      Only demo, but you can buy codec online (100$). Who test it?
      http://www.leadcodecs.com/Codecs/LEAD-H264.htm
      ---
  10. Compression Master 3 (Popwire/Teleca AB)
      H264 encoder is all a part of Popwire's Compression Master.
      Demo soft available (encoding is limited to 20 seconds).
      http://www.popwire.com/product_info.php?cPath=1&products_id=2
      ---
  11. SVM H.264 Decoder Kit & MKi DVD Converter (Pegasus Information Technology Inc.)
      H.264 encoder and decoder with non standart format.
      http://www.h-264.com/downloads.htm
      ---
  12. AVC Alliance free demo player (AVC Alliance, written by Philips Electronics)
      Poor but free.
      http://www.avc-alliance.nl/main/downloads.htm
      ---
  13. Intel IPP H.264 codec (Intel Corporation)
      Intel IPP Library containe now H.264 encoder and decoder.
      Higher quality codecs are promised in 5.0 version (released now)
      http://www.intel.com/cd/software/products/asmo-na/eng/perflib/ipp/index.htm
      ---
  14. PowerEncoder MPEG-4 AVC (CyberLink Corp.) NEW!!!
      Available in online shop.
      http://www.cyberlink.com/multi/products/main_43_ENU.html
      ---
  15. ATI H.264 (ATI Technologies Inc.)
      No download. Software only and +HW support. Very fast codec by our tests.
      http://www.ati.com/technology/h264.html
      ---
  16. Mpegable AVC Codec (dicas digital image coding GmbH)
      No download, VfW
      http://www.mpegable.com/show/mpegableavc.html
      Old version from free-codecs:
      http://www.codec-download.com/modules.php?name=Downloads&d_op=getit&lid=436
      ---
  17. Expert H.264 (PixelTools Corporation)
      Download on request
      http://www.pixeltools.com/experth264.html
      ---
  18. Softstream H.264/MPEG-4 (Media Excel)
      Audio & Video encoders ane decoders. Evaluation copy on request
      http://www.mediaexcel.com/products.htm
      ---
  19. MPEG-2/HDV/H.264 software (KDDI R&D Labs. Inc.)
      No download. MPEG related products also with H.264 support
      http://avs.kddilabs.jp/mpeg/indexe.html
      ---
  20. Fraunhofer IIS H.264 Codec (Fraunhofer IIS)
      No download.
      http://www.iis.fraunhofer.de/amm/download/mpeg4/
      ---
  21. UBLive-264-C64 (UB Video Incorporated)
      Demo available on request.
      http://www.ubvideo.com/mainmenu.html
      ---
  22. Sorenson Squeeze 4 Compression Suite (Sorenson)
      No download. Converter.
      http://www.sorensonmedia.com/solutions/prod/mpg4_win.php
      ---
  23. Sonic?s HD-Series AVC encoder (Sonic Solutions)
      No download. Only press release now.
      http://www.sonic.com/about/press/news/2005/04/hdseries.aspx
      ---
  24. Enchansed sklmp4 (Pascal Massimino)
      No download. Announce of 264 development.
      http://skal.planet-d.net/coding/mpeg4codec.html
      ---
  25. NEX VISION H.264 (NEX VISION)
      No download.
      http://www.nexvision.fr/
      ---
  26. Hughes Network Systems H.264 (Hughes Network Systems, LLC)
      No download. Own Win&Linux codec for internal usage
      http://www.hns.com/HNS/Doc/0/MGD6UC9UN984BDH4VOK3N53S37/03-03-04_IPoS.html
      ---
  27. Imagicity H.264/AVC codec (Imagicity, UK) NEW!!!
      No download. Speed optimization claimed.
      http://www.imagicity.com/products/products.htm
      ---
      ----------------------------------------------------------
      OPEN SOURCE:
      ---------------------
  28. x264 (free, supported by VideoLAN)
      OPEN SOURCE. Good publically available high profile encoder.
      Developers declare "early development stage", but encode results are one of the best.
      http://www.videolan.org/x264.html
      http://x264.nl/
      ---
  29. JM 9.6 reference H.264 encodec (free, supported by Joint Video Team ISO+ITU-T)
      OPEN SOURCE. Can be used only for stream testing.
      http://bs.hhi.de/~suehring/tml/
      ---
  30. libavcodec H.264 Decoder (Open Source Project FFMPEG)
      OPEN SOURCE. Permanently updated decoder.
      http://ffmpeg.sourceforge.net/index.php
      ---
  31. Hdot264 Pre-Alpha (Open Source Project Hdot264)
      OPEN SOURCE.
      http://sourceforge.net/projects/hdot264/
      ---
  32. Minimal H.264 decoder (by Martin Fiedler)
      OPEN SOURCE. Basic H.264 decoder.
      http://www-user.tu-chemnitz.de/~mfie/stuff/SA/
      ---
      See also Software Developments Kits for software codec usage:
      ---
         1. Elecard AVC&MPEG2 Encoder SDK (Elecard)
            MPEG2 and H.264 Video Encoder and Decoders other
            components within the Microsoft? DirectShow? technology.
            http://www.elecard.com/products/encoder_sdk.shtml
            http://www.elecard.com/products/decoder_sdk.shtml
         2. MainConcept H.264 SDK (MainConcept AG)
            Windows, Apple, Linux and TI DSP supported.
            http://developer.mainconcept.com/h264-sdk.html
         3. LEADTOOLS Multimedia Developer SDKs (LEAD Technologies, Inc)
            Support for many-many formats, including H.264
            http://www.leadtools.com/SDK/Multimedia/Multimedia-Products-n.htm
         4. MPEG SDK "MP-Factory" (KDDI R&D Labs. Inc.)
            SDK with H.264, HDV, MPEG-1,-2, and -4 support
            http://avs.kddilabs.jp/mpeg/mpfs32/indexe.html
         5. VSS H.264 SDK (Vanguard Software Solutions, Inc.)
            Based on there codec. Windows and Linux support.
            http://www.vsofts.com/h264/sdk.html
         6. Pegasus H.264 SDK Development Kits (Pegasus)
            Looks like simple codec, but prefilters declared.
            http://www.h-264.com/Products.htm
      ----------------------------------------------------------
      HARDWARE IMPLEMENTED:
      ---------------------
  33. Sony PlayStation Portable (Sony Computer Entertainment Inc.)
      Product: Support H.264/MPEG-4 AVC Main Profile Level3 decoding.
      http://www.us.playstation.com/consoles.aspx?id=4
      See also Image Converter 2 Plus:
      http://www.jp.sonystyle.com/Nws/Software_dl/Pc/Software/Haa/2105710489100.html
      ---
  34. VSX 300 Videoconferencing Terminal with H.264 support (Polycom Inc)
      Product: Biggest HW videoconferencing company, prepare many solutions with 264 support
      http://www.polycom.com/products_services/1,1443,pw-35-6197,00.html
      ---
  35. Tandberg EN5930 (Tandberg Television)
      Product for videoconferencing. Who test it? I have bad review.
      http://www.tandbergtv.com/productview.asp?n=65
      ---
  36. Tandberg Telecom - video communication (Tandberg Telecom)
      Product for videoconferencing, supports H.264 baseline profile.
      http://www.tandberg.net/collateral/white_papers/H264.pdf
      ---
  37. 4Caster MPEG-4 / H.264 Encoder (Envivio, Inc.)
      Product: Hardware box encoder
      http://www.envivio.com/products/4caster.html
      http://www.envivio.com/products/4caster_se.html (software, no download)
      ---
  38. Tut Systems MPEG-4 AVC (Tut Systems, Inc.)
      Product: Hardware transcoder box.
      http://www.tutsys.com/pdflibrary/pdf/MPEG4_AVC_2_9_05.pdf
      ---
  39. Hardware MGW 5100 (Optibase, Inc.)
      Product: Hardware box encoder
      http://www.optibase.com/Content.aspx?id=38
      ---
  40. Harmonic DiviCom? MV 100 Encoder (Harmonic Inc. )
      Product: Hardware box encoder
      http://www.harmonicinc.com/view_csd_product_group.cfm?classID=1010
      ---
  41. SkyStream Mediaplex & iPlex (SkyStream Networks , inc.)
      Product: Hardware box encoder.
      Platforms contain the AVC Encoder Submodule, press release only
      http://www.skystream.com/press/2005/041805-mpeg4.asp
      ---
  42. E42 - H264 DVB over IP Broadcast Encoder (StreamTel Corporation)
      Product: Hardware box encoder.
      Broadcast-Grade real time H264 DVB IP streaming encoder
      http://www.streamtel.com/streamtel/products/det/1_Encoders/46_E42_-_H264_DVB_over_IP_Broadcast_Encoder.htm
      ---
  43. Darim MV401EXP-H264 Encoder and MV410EXP Decoder (Darim Vision Co., Ltd.)
      Product: Hardware box Encoding/Streaming systems
      http://www.darimmedia.com/products/mv401exp-h264.php
      ---
  44. Mayah I [io] 8000 series (MAYAH Communications)
      Product: Hardware box. Plan AVC availability in Q3 '05.
      http://www.mayah.com/products/io-8000a.htm
      ---
  45. Thales ARGOS MPEG-4 Encoder (Thales Broadcast & Multimedia, Inc.)
      Product: Hardware box. Plan AVC from July 05.
      http://www-db.thomcastcom.ext.coltfrance.com/db/thomcast/webdriver?MIval=V_Prod_Detail_b.html&F_ID=4&D_ID=1&P_ID=221
      ---
  46. ViBE MPEG-4 AVC (Thomson)
      Chip: modular hardware series of encoders
      http://www.thomsongrassvalley.com/products/transmission/vibe/encoder_mpeg4/
      ---
  47. Broadcom chips (Broadcom Corp. former Sandvideo)
      Chip with H.264 HD / decoding only
      http://www.broadcom.com/products/Cable/HDTV-SDTV-Video,-Graphics-%26-Receiver-Chips/BCM7411
      ---
  48. Sentivision H.264/AVC Decoder (Sentivision, Japan)
      Chip with H.264 HD / decoding only / only for TI DSP's
      http://www.sentivision.com/products/h264/h264-en.html
      ---
  49. SA AVC Encoder (Scientific Atlanta)
      Product: Transmission system (for satellite TV)
      http://www.saeurope.com/products/customerhome.htm
      ---
  50. H.264 VLE4000 (LSI Logic Corp. former: VideoLocus)
      Chips with 264 support
      http://www.lsilogic.com/products/video_broadcasting/vle4000.html
      ---
  51. Modulus Video SDTV, HDTV Encoder (Modulus Video, Inc.)
      Chips. Declarations only now. License H.264 from LSI Logic.
      http://modulusvideo.com/main.php?Page=12
      ---
  52. BSP 15 (System-on-chip) (Equator Technologies, Inc.)
      Chips. No exact codec info.
      http://www.equator.com/solutions/digitalmedia.html
      ---
  53. BT Exact H.264 Suite (BT Exact)
      temporarily not available
      http://www.btexact.com/technologies/ipprofiles?doc=42895
      ---
  54. Surveillance Real Time H.264 Encoders (DSP Research, Inc.)
      Several chips implementation.
      http://www.dspr.com/www/products/overview_video.htm
      http://www.dspr.com/www/support/download/video_download.htm
      ---
  55. VP3-PMC / Hardware TwinPEG Pro (Vitec Multimedia)
      Chips. TMS320DM642TMDSP & etc
      http://www.vitecmm.com/product.php?id=41
      ---
  56. WWComs H.264 Codecs (W&W Communications Inc.)
      Several chips implementation (TI DM64x, C64xx, Freescale DSPs).
      http://www.wwcoms.com/products/h264_codec.htm
      http://www.wwcoms.com/products/codec/bc264.htm
      ---
  57. CX2418x H.264 Video Decoder IC (Conexant Systems, Inc.)
      HW Decoder with Main & High Profile (on CX24182)
      http://www.conexant.com/products/entry.jsp?id=920
      ---
  58. Sigma Designs HD decoder (Sigma Designs, Inc.)
      HW HD Decoder based on SMP8630 family
      http://www.sigmadesigns.com/products/SMP8630series.htm
      ---
  59. Cradle Technologies H.264 (Cradle Technologies, Inc.)
      HW Main profile Encoder based on CT3600 MDSP family
      http://www.cradle.com/products/asw_ct3600_h264.shtml
      ---
  60. STMicroelectronics H.264 (STMicroelectronics)
      HW HD High Profile decoder
      http://www.st.com/stonline/products/literature/bd/11102.htm
      ---
  61. PacketVideo H.264 (PacketVideo Corp.)
      HW Base Profile encoder/decoder on ARM/OMAP/DM642
      http://www.packetvideo.com/solutions/
      ---
   We plan prepare next MSU H.264 codecs comparison test soon will be
   available from
   http://compression.ru/video/codec_comparison/mpeg-4_avc_h264_en.html
   If you know any more information about H.264 codecs, please reply
   this message or add information to Compression-links.info H.264 page:
   http://www.compression-links.info/MPEG-4_AVC_H264
============================================================================
-- 
Best regards,
 Dmitriy Vatolin 
  Ph.D, Head of Video Group                        
   mailto:dmitriy@graphics.cs.msu.ru
From may_ank77  yahoo.com  Thu Nov 10 21:04:40 2005
From: may_ank77  yahoo.com (mayank agarwal)
Date: Fri Nov 11 16:00:39 2005
Subject: [Mp4-tech] Iprof --Complexity analysis
In-Reply-To: <139c4c630511092354p28762ecer89bf38e786213ec0@mail.gmail.com>
Message-ID: <20051111050440.95695.qmail@web52810.mail.yahoo.com>
Please forward the reply of this mail to me also.
I also want that.Also if some one can clarify the
following things then it will be of great help.
1.Is it possible to implement motion estimation can be
implemented as standalone application upon any image
which we can see in YUV viewer and then analyse the
results and based upon that we can deduce that if it
is made a part of any algorithm like H.264/MPEG4 then
what will be the performance in terms of frame/s and
bit rate.
2.How do we judge the suitability of particular motion
estimation algorithm.
3.In the JM reference code for H.264 a combination of
integer pixel motion estimation and fractional pixel
estimation is used and rate distortion optimization
is also used.My questins are:
   1.If there are any documents which explain in
detail how to do the above things.
   2.If any real time implementation of H.264 supports
that kind of technique and the performance results
from that.
--- Humaira <humaira.nisar@gmail.com> wrote:
> Hi
>  Do anyone has an idea how to get iprof. Or is there
> some other good
> software that helps in the complexity analysis of
> motion estimation
> algorithms.
> Also i would like to know about some place that will
> help in the programming
> of motion estimation algorithms
>  Thanks
> > _______________________________________________
> NOTE: Please use clear subject lines for your posts.
> Include [audio, [video], [systems], [general] or
> another apppropriate identifier to indicate the type
> of question you have.
> 
> Note: Conduct on the mailing list is subject to the
> Antitrust guidelines found at
http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
__________________________________ 
Yahoo! Mail - PC Magazine Editors' Choice 2005 
http://mail.yahoo.com
From smikkyssingh  gmail.com  Fri Nov 11 16:59:54 2005
From: smikkyssingh  gmail.com (Smikky Sahaya Singh)
Date: Fri Nov 11 16:00:44 2005
Subject: [Mp4-tech] .mp4 file format
Message-ID: <2d0351010511110329l4d6c6a5p843daeff30442ef8@mail.gmail.com>
Hi,
 How to extract the video and audio part from the .mp4 container?
 Is there any document which explains the .mp4 file format(Explaining the
Packing of Video and Audio)?
 Thanks in Advance.
JC
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051111/8b89d664/attachment.html
From may_ank77  yahoo.com  Fri Nov 11 04:13:15 2005
From: may_ank77  yahoo.com (mayank agarwal)
Date: Fri Nov 11 16:00:49 2005
Subject: [Mp4-tech] Some doubts in video
Message-ID: <20051111121315.70547.qmail@web52810.mail.yahoo.com>
Hi all, 
 Recently some person asked me some questions about
video.I found myself cluless on those questions.
Please help me to clarify on the following topics.
1.In MPEG-4 simple profile how many motion vectors are
transmitted from encoder to decoder or generalizing
the above can anybody tell how to determine the number
of motion vectors being transmitted from encoder to
decoder.
2.Suppose in the video encoding we don't do
quantization then what will be affect on final encoded
stream.
3.Suppose we are encoding QCIF sized video then what
will be the minimum size of decoded buffer.I suppose
it is 176*144+2*176/2*144/2.Please correct me if i am
wrong.
4.Suppose we are doing some video integration and we
found out that DSP memory is not enough for the
encoded stream then how can we increase DSP memory.
Thanks and Regards,
Mayank
__________________________________ 
Yahoo! FareChase: Search multiple travel sites in one click.
http://farechase.yahoo.com
From singer  apple.com  Fri Nov 11 10:23:49 2005
From: singer  apple.com (Dave Singer)
Date: Mon Nov 14 12:18:43 2005
Subject: [Mp4-tech] .mp4 file format
In-Reply-To: <2d0351010511110329l4d6c6a5p843daeff30442ef8@mail.gmail.com>
References: <2d0351010511110329l4d6c6a5p843daeff30442ef8@mail.gmail.com>
Message-ID: <p06230912bf9a91f572fe@[17.219.157.83]>
At 16:59  +0530 11/11/05, Smikky Sahaya Singh wrote:
>Hi,
>
>How to extract the video and audio part from the .mp4 container?
>
>Is there any document which explains the .mp4 file format(Explaining 
>the Packing of Video and Audio)?
ISO/IEC 14496-14 (mpeg-4 specific handling), ISO/IEC 14496-12 (all 
the structure)
   part 12 is freely available from ISO
you might also look at
ISMA 1.0 and ISMA 2.0, which are also free.
>
>Thanks in Advance.
>JC
>
>
>_______________________________________________
>NOTE: Please use clear subject lines for your posts. Include [audio, 
>[video], [systems], [general] or another apppropriate identifier to 
>indicate the type of question you have.
>
>Note: Conduct on the mailing list is subject to the Antitrust 
>guidelines found at 
>http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-- 
David Singer
Apple Computer/QuickTime
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051111/ccfb3c7a/attachment.html
From alexis.tourapis  dolby.com  Fri Nov 11 12:17:41 2005
From: alexis.tourapis  dolby.com (Tourapis, Alexis)
Date: Mon Nov 14 12:26:27 2005
Subject: [Mp4-tech] Iprof --Complexity analysis
Message-ID: <7272EE229DA1AA48B47EBDC47EB0C68B019D9167@sapphire.dolby.net>
Dear Humaira and Mayank,
I am not sure if Iprof is still available or supported anymore. Last time that I had checked the official website (i.e. http://www.lis.ei.tum.de/research/bv/topics/method/e_iprof.html ) was providing a dead link. 
However, as many people would tell you, you may instead use many free or commercial profiler applications. You may wish to check for example gprof, vtune, codeanalys, the devpartner profiler. Note that there is also another profiler named iprof at : 
http://silverspaceship.com/src/iprof/
but this seems to have no relationship with the original. 
You may also want to check the atomium suite at http://www.imec.be/design/atomium/toolsuite.shtml <http://www.imec.be/design/atomium/toolsuite.shtml>  while, some nice work (and obvious quite relevant) has also been done from EPFL. I would suggest getting hold of their MPEG M9553 contribution, named "Complexity Evaluation of Different Configurations of the JVT Codec". In fact you may download their full contribution also from the JVT site (see http://ftp3.itu.ch/av-arch/jvt-site/2003_03_Pattaya/, contribution JVT-G038, which also includes their software). 
With regards to some of the questions on motion estimation, motion estimation does not need to be only part of an encoding solution. Motion estimation is in fact used in many different systems for various purposes ranging from pre and post filtering, segmentation, registration, scene analysis etc. Therefore, the answer is yes you can implement motion estimation as a standalone module. However (and this is a big however), one thing you need to consider is the fact that within an encoder you actually introduce artifacts due to the encoding process itself which would have an immediate impact on the ME process as well (assuming that you consider reconstructed images as references). Note also that a video codec, such as H.264, contains also several other components that would impact motion estimation, such as Rate Distortion based optimized mode decision, Rate control etc which will also affect your decisions and your final motion field. Many algorithms also at times consider motion fields of previously coded pictures to reduce complexity which would require you a bit more work to add in an outside tool (i.e. you would need to create a buffer as well). 
Your second question is maybe a bit tough to answer considering that the answer depends on many parameters. I.e. is your implementation a hardware or software one, what constraints do you have in terms of memory and processing power, can you use parallelization instructions or not, do you have issues with branching etc. 
You may wish to check JVT-O079 for information on some of the schemes the JM is using. Note though that this document is currently being updated and hopefully there will be a much better version before the next meeting describing more accurately the software's behavior and the algorithms available in the software. Note that we are also trying to rearchitecture somewhat the JM encoder and decoder as to make them faster and a bit more friendlier.
Best regards,
Alexis
	-----Original Message----- 
	From: mp4-tech-bounces@lists.mpegif.org on behalf of mayank agarwal 
	Sent: Thu 11/10/2023 9:04 PM 
	To: Humaira; mp4-tech@lists.mpegif.org 
	Cc: 
	Subject: Re: [Mp4-tech] Iprof --Complexity analysis
	Please forward the reply of this mail to me also.
	I also want that.Also if some one can clarify the
	following things then it will be of great help.
	1.Is it possible to implement motion estimation can be
	implemented as standalone application upon any image
	which we can see in YUV viewer and then analyse the
	results and based upon that we can deduce that if it
	is made a part of any algorithm like H.264/MPEG4 then
	what will be the performance in terms of frame/s and
	bit rate.
	2.How do we judge the suitability of particular motion
	estimation algorithm.
	3.In the JM reference code for H.264 a combination of
	integer pixel motion estimation and fractional pixel
	estimation is used and rate distortion optimization
	is also used.My questins are:
	   1.If there are any documents which explain in
	detail how to do the above things.
	   2.If any real time implementation of H.264 supports
	that kind of technique and the performance results
	from that.
	--- Humaira <humaira.nisar@gmail.com> wrote:
	> Hi
	>  Do anyone has an idea how to get iprof. Or is there
	> some other good
	> software that helps in the complexity analysis of
	> motion estimation
	> algorithms.
	> Also i would like to know about some place that will
	> help in the programming
	> of motion estimation algorithms
	>  Thanks
	> > _______________________________________________
	> NOTE: Please use clear subject lines for your posts.
	> Include [audio, [video], [systems], [general] or
	> another apppropriate identifier to indicate the type
	> of question you have.
	>
	> Note: Conduct on the mailing list is subject to the
	> Antitrust guidelines found at
	http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
	__________________________________
	Yahoo! Mail - PC Magazine Editors' Choice 2005
	http://mail.yahoo.com
	_______________________________________________
	NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have.
	Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-----------------------------------------
This message (including any attachments) may contain confidential
information intended for a specific individual and purpose.  If you are
not the intended recipient, delete this message.  If you are not the
intended recipient, disclosing, copying, distributing, or taking any
action based on this message is strictly prohibited.
From eng.mjubran  gmail.com  Sat Nov 12 21:10:37 2005
From: eng.mjubran  gmail.com (Mohammad Jubran)
Date: Mon Nov 14 12:30:35 2005
Subject: [Mp4-tech] duplicate frame_num in short-term reference picture
	buffer
Message-ID: <acb31d960511121810s351506dbt44d9d1883992805e@mail.gmail.com>
Hello
 I am trying to test the optimality of H264 in wireless systesm. Droping
some partitions give the following message while decoding :
 Flush remaining frames from dpb. dpb.size=16, dpb.used_size=4
output frame with frame_num #0, poc 0 (dpb. dpb.size=16, dpb.used_size=4)
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=3
output frame with frame_num #2, poc 2 (dpb. dpb.size=16, dpb.used_size=3)
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=2
output frame with frame_num #1, poc 4 (dpb. dpb.size=16, dpb.used_size=2)
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=1
output frame with frame_num #2, poc 8 (dpb. dpb.size=16, dpb.used_size=1)
duplicate frame_num in short-term reference picture buffer
 Can Anyone help me figure out why is that happening?
--
Regards
Mohammad Khalil Issa Jubran
State University Of New York at Buffalo "SUNY at Buffalo"
School of Electrical Engineering
PhD. Student.
Tel. 001-716-400-4677
Mailing Address
123 Hawthorne ave "upper"
Tonawanda - NY - 14223 , USA
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051112/a6de3d43/attachment.html
From eng.mjubran  gmail.com  Sat Nov 12 21:23:31 2005
From: eng.mjubran  gmail.com (Mohammad Jubran)
Date: Mon Nov 14 12:30:41 2005
Subject: [Mp4-tech] problem with H.263 codec version JM10.1 "duplicate
	frame_num in short-term reference picture buffer "
Message-ID: <acb31d960511121823y3790aba3l1c22baac078b7884@mail.gmail.com>
Hello
 *I am using H.264 codec version JM 10.1 and I have encoded my **video in
the RTP packet format .DP is enabled. I use some packet dropping scheme to *
*drop the RTP packets (packet is a partition of a slice) as per a given
Packet Loss Rate and feed the rest **to the decoder. But for any case, I get
an error which says
**
**Flush remaining frames from dpb. dpb.size=16, dpb.used_size=4
output frame with frame_num #0, poc 0 (dpb. dpb.size=16, dpb.used_size=4)
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=3
output frame with frame_num #2, poc 2 (dpb. dpb.size=16, dpb.used_size=3)
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=2
output frame with frame_num #1, poc 4 (dpb. dpb.size=16, dpb.used_size=2)
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=1
output frame with frame_num #2, poc 8 (dpb. dpb.size=16, dpb.used_size=1)
duplicate frame_num in short-term reference picture buffer
**
**and the decoder stops. I am not sure why this happens. Can anyone **please
explain the reason and what can be done in this regard. It is **really
urgent for me and so I request you to kindly help me out with **your
suggestions.....
*
--
Thanks and Regards
Mohammad Khalil Issa Jubran
State University Of New York at Buffalo "SUNY at Buffalo"
School of Electrical Engineering
PhD. Student.
Tel. 001-716-400-4677
Mailing Address
123 Hawthorne ave "upper"
Tonawanda - NY - 14223 , USA
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051112/4141e459/attachment.html
From china264  gmail.com  Mon Nov 14 14:24:03 2005
From: china264  gmail.com (China264)
Date: Mon Nov 14 12:30:46 2005
Subject: [Mp4-tech] about init reference list
Message-ID: <ac4cd46a0511132124o6955f1dqd33bfae2a809dae5@mail.gmail.com>
Dear all,
 In " Initialisation process for reference picture lists for B slices in
fields", there are a sentence as following,
 When the reference picture list RefPicList1 has more than one entry and it
is identical to the reference picture list RefPicList0, the first two
entries RefPicList1[0] and RefPicList1[1] are switched.
I have checked the source code of JM 86. The source code do it like this.
 Can some expert tell me why do like this?
 thanks!
 jeff
--
-----------------------------------------------
Video chip design BBS
China264.68L.com <http://China264.68L.com>
China264@gmail.com
------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051114/9ef6db9f/attachment.html
From tma  iis.fhg.de  Mon Nov 14 12:31:24 2005
From: tma  iis.fhg.de (Herbert Thoma)
Date: Mon Nov 14 12:30:52 2005
Subject: [Mp4-tech] Some doubts in video
In-Reply-To: <20051111121315.70547.qmail@web52810.mail.yahoo.com>
References: <20051111121315.70547.qmail@web52810.mail.yahoo.com>
Message-ID: <4378758C.8070809@iis.fhg.de>
mayank agarwal wrote:
> Hi all, 
>  Recently some person asked me some questions about
> video.I found myself cluless on those questions.
> Please help me to clarify on the following topics.
> 
> 1.In MPEG-4 simple profile how many motion vectors are
> transmitted from encoder to decoder or generalizing
> the above can anybody tell how to determine the number
> of motion vectors being transmitted from encoder to
> decoder.
Hmm, decode the bitstream? The extreme cases are:
- no motion vectors at all (e.g. IVOPs)
- all macroblocks coded with 4 vectors (this is
   probably not very likely)
There is a mechanism called "complexity estimation"
in the standard but as far as I know nobody uses this ...
> 2.Suppose in the video encoding we don't do
> quantization then what will be affect on final encoded
> stream.
That's not allowed by the standard. So your stream
will not be valid.
> 3.Suppose we are encoding QCIF sized video then what
> will be the minimum size of decoded buffer.I suppose
> it is 176*144+2*176/2*144/2.Please correct me if i am
> wrong.
That's correct.
> 4.Suppose we are doing some video integration and we
> found out that DSP memory is not enough for the
> encoded stream then how can we increase DSP memory.
I don't understand this question. Add more memory to
the DSP?
Kind regards,
  Herbert.
> Thanks and Regards,
> Mayank
> 
> 
> 		
> __________________________________ 
> Yahoo! FareChase: Search multiple travel sites in one click.
> http://farechase.yahoo.com
> _______________________________________________
> NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have.
> 
> Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
> 
-- 
Herbert Thoma
Head of Video Group
Multimedia Realtime Systems Department
Fraunhofer IIS
Am Wolfsmantel 33, 91058 Erlangen, Germany
Phone: +49-9131-776-323
Fax:   +49-9131-776-399
email: tma@iis.fhg.de
www: http://www.iis.fhg.de/
From Ilan.Daniel  Adimos.com  Mon Nov 14 13:48:25 2005
From: Ilan.Daniel  Adimos.com (Ilan Daniel)
Date: Mon Nov 14 12:30:58 2005
Subject: [Mp4-tech] GOP concept in H.264
Message-ID: <2A01AFD68E343242BA2C748A5E1A733D9D9D52@jerry.Adimos.com>
Gary,
 thanks for the link, it is long time I was needing this ?reference software manual? 
I did not know it exists?
tummala kalyani,
I also faced same problem when I set NumberBFrames to more than 0,
and when I set FrameSkip = NumberBFrames, I got the desirable results.
Note that in that case you need to re calculate the number of FramesToBeEncoded
And now I see it in the manual?..
Regards,
Ilan
________________________________
From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Gary Sullivan
Sent: ? 09 ?????? 2005 20:54
To: tummala kalyani; mp4-tech@lists.mpegif.org
Subject: RE: [Mp4-tech] GOP concept in H.264
I suggest reading the reference software manual found as document JVT-Q042 in the 2005_10_Nice directory on the JVT document site http://ftp3.itu.int/av-arch/jvt-site.
The standard itself does not actually include a definition of a concept called a GOP.  When people refer to "open GOP", they are usually talking about non-reference pictures that occur in the bitstream after an Intra picture that are displayed prior to the Intra picture being allowed to use pictures that precede the Intra picture in decoding order as references for their motion compensation prediction decoding process.
You sound a bit confused about what the standard does, or perhaps it is just a question of phrasing.  I must warn you that the reference software is really meant for those who have a significant amount of familiarity with video compression technology.  Regarding your confusion over why the JM reference software might use more bits to code a video clip than some MPEG-2 reference software, could it be that the two encoded video clips do not have comparable decoded picture quality?  Targeting equal output video quality is a fundamental part of claims of relative bit rate savings performance capability.
Best Regards,
Gary Sullivan
________________________________
	From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of tummala kalyani
	Sent: Tuesday, November 08, 2023 8:45 AM
	To: mp4-tech@lists.mpegif.org
	Subject: [Mp4-tech] GOP concept in H.264
	Hi All,
	I would like to know about GOP concept in H.264. In JM encoder config file there is a parameter "support for open GOP". How can I use this? I tried to 1-enable and 0-disable options. But I don't find any difference. 
	There is provision to set I frame distance and number of B frames. When I treid to set number of B frames more than 0 , it gives me an error message saying that  "Number of B frames should me more than frames skipped in the input". What is the use of Frames skipped parameter in the encoder.cfg file in JM?
	I need bitrate vs psnr comparison graphs for H.264 codec and MPEG-2 codec. For that I am running JM and MPEG-2 reference softwares with foreman.yuv as input video. I found that for baseline and main profiles H.264 codec output file size  is double to MPEG-2 file size. I use this filesize to calculate bitrate( filesize * (frame rate / total frames)). But according to specs. H.264 codec bitrate should be half to that of MPEG-2. But I am getting in reverse. I am confused by this. I think I have not understood it properly. Can some one explain me this clearly?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051114/bf16aa0f/attachment.html
From giri_tammana  yahoo.co.in  Mon Nov 14 18:17:22 2005
From: giri_tammana  yahoo.co.in (Giridhar)
Date: Mon Nov 14 21:11:32 2005
Subject: [Mp4-tech] H264 Deblocking
Message-ID: <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA7QqDKrLZ/keAzZXjHS06VMKAAAAQAAAAYYRZtXjRL0+UyfbMIpzsYwEAAAAA@yahoo.co.in>
Hello all,
My question is filter decision, as deblocking is done for 4x4 block
boundaries in a MacroBlock do the values of alpha and beta vary for each 4x4
block boundary or will it be constant for entire MB?
TIA
giridhar
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051114/9a8dbbf3/attachment.html
From lahmurmusic  mac.com  Mon Nov 14 20:36:09 2005
From: lahmurmusic  mac.com (Joseph Lahart)
Date: Mon Nov 14 21:19:00 2005
Subject: [Mp4-tech] Audio /Video   Podcast.
Message-ID: <08f4ab0fe511938f47974bccf0d72688@mac.com>
Hi my first time here.
I have .mac account and want to up load this to a pod cast. I have red 
some articals and gone to various web sites but
am no closer. There is plenty of info about audio but i have audio and 
video. Any help would be appreciated.
Rgds
Joseph Lahart
From garysull  windows.microsoft.com  Mon Nov 14 13:21:41 2005
From: garysull  windows.microsoft.com (Gary Sullivan)
Date: Tue Nov 15 08:32:16 2005
Subject: [Mp4-tech] Some doubts in video
Message-ID: <91D7F2CEE3425A4A9D11311D09FCE246118627CC@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
Mayank, Herbert, et al,
RE:
+> > 2.Suppose in the video encoding we don't do
+> > quantization then what will be affect on final encoded
+> > stream.
+> 
+> That's not allowed by the standard. So your stream
+> will not be valid.
Strictly speaking, I disagree.  The standard does not specify how
encoders must operate.  It only specifies constraints on the format of
bitstreams and the results of the decoding process for such bitstreams
that obey those constraints.
If you can think of how to design an encoder that obeys all constraints
in the standard without that encoder performing "quantization", then
please feel free to go right ahead.
Best Regards,
Gary Sullivan
+> -----Original Message-----
+> From: mp4-tech-bounces@lists.mpegif.org 
+> [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Herbert Thoma
+> Sent: Monday, November 14, 2023 3:31 AM
+> To: mayank agarwal
+> Cc: mp4-tech@lists.mpegif.org; c6x@yahoogroups.com
+> Subject: Re: [Mp4-tech] Some doubts in video
+> 
+> mayank agarwal wrote:
+> > Hi all, 
+> >  Recently some person asked me some questions about
+> > video.I found myself cluless on those questions.
+> > Please help me to clarify on the following topics.
+> > 
+> > 1.In MPEG-4 simple profile how many motion vectors are
+> > transmitted from encoder to decoder or generalizing
+> > the above can anybody tell how to determine the number
+> > of motion vectors being transmitted from encoder to
+> > decoder.
+> 
+> Hmm, decode the bitstream? The extreme cases are:
+> - no motion vectors at all (e.g. IVOPs)
+> - all macroblocks coded with 4 vectors (this is
+>    probably not very likely)
+> 
+> There is a mechanism called "complexity estimation"
+> in the standard but as far as I know nobody uses this ...
+> 
+> > 2.Suppose in the video encoding we don't do
+> > quantization then what will be affect on final encoded
+> > stream.
+> 
+> That's not allowed by the standard. So your stream
+> will not be valid.
+> 
+> > 3.Suppose we are encoding QCIF sized video then what
+> > will be the minimum size of decoded buffer.I suppose
+> > it is 176*144+2*176/2*144/2.Please correct me if i am
+> > wrong.
+> 
+> That's correct.
+> 
+> > 4.Suppose we are doing some video integration and we
+> > found out that DSP memory is not enough for the
+> > encoded stream then how can we increase DSP memory.
+> 
+> I don't understand this question. Add more memory to
+> the DSP?
+> 
+> Kind regards,
+>   Herbert.
+> 
+> > Thanks and Regards,
+> > Mayank
+> > 
+> > 
+> > 		
+> > __________________________________ 
+> > Yahoo! FareChase: Search multiple travel sites in one click.
+> > http://farechase.yahoo.com
+> > _______________________________________________
+> > NOTE: Please use clear subject lines for your posts. 
+> Include [audio, [video], [systems], [general] or another 
+> apppropriate identifier to indicate the type of question you have.
+> > 
+> > Note: Conduct on the mailing list is subject to the 
+> Antitrust guidelines found at 
+> http://www.mpegif.org/public/documents/vault/mp-out-30042-Ant
+> itrust.php
+> > 
+> 
+> -- 
+> Herbert Thoma
+> Head of Video Group
+> Multimedia Realtime Systems Department
+> Fraunhofer IIS
+> Am Wolfsmantel 33, 91058 Erlangen, Germany
+> Phone: +49-9131-776-323
+> Fax:   +49-9131-776-399
+> email: tma@iis.fhg.de
+> www: http://www.iis.fhg.de/
+> _______________________________________________
+> NOTE: Please use clear subject lines for your posts. Include 
+> [audio, [video], [systems], [general] or another 
+> apppropriate identifier to indicate the type of question you have.
+> 
+> Note: Conduct on the mailing list is subject to the 
+> Antitrust guidelines found at 
+> http://www.mpegif.org/public/documents/vault/mp-out-30042-Ant
+> itrust.php
+> 
From garysull  windows.microsoft.com  Mon Nov 14 13:23:05 2005
From: garysull  windows.microsoft.com (Gary Sullivan)
Date: Tue Nov 15 08:32:22 2005
Subject: [Mp4-tech] problem with H.263 codec version JM10.1
	"duplicateframe_num in short-term reference picture buffer "
Message-ID: <91D7F2CEE3425A4A9D11311D09FCE246118627D7@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
Mohammad et al,
Probably one of the best people to talk to about this is your
SUNY-Buffalo colleage Saurav Bandyopadhyay [skb3@buffalo.edu].
Graceful behavior of a decoder under packet loss conditions is a
desirable, but not mandatory, feature for implementations (from the
standardization perspective, that is -- perhaps not from the
marketability perspective).
Best Regards,
Gary Sullivan
________________________________
	From: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Mohammad Jubran
	Sent: Saturday, November 12, 2023 6:24 PM
	To: mp4-tech@lists.mpegif.org; eng.mjubran@gmail.com
	Subject: [Mp4-tech] problem with H.263 codec version JM10.1
"duplicateframe_num in short-term reference picture buffer "
	Hello
	I am using H.264 codec version JM 10.1 and I have encoded my
video in the RTP packet format .DP is enabled. I use some packet
dropping scheme to drop the RTP packets (packet is a partition of a
slice) as per a given Packet Loss Rate and feed the rest to the decoder.
But for any case, I get an error which says
	Flush remaining frames from dpb. dpb.size=16, dpb.used_size=4 
	output frame with frame_num #0, poc 0 (dpb. dpb.size=16,
dpb.used_size=4 ) 
	Flush remaining frames from dpb. dpb.size=16, dpb.used_size=3 
	output frame with frame_num #2, poc 2 (dpb. dpb.size=16,
dpb.used_size=3) 
	Flush remaining frames from dpb. dpb.size=16, dpb.used_size=2 
	output frame with frame_num #1, poc 4 (dpb. dpb.size=16,
dpb.used_size=2) 
	Flush remaining frames from dpb. dpb.size=16, dpb.used_size=1 
	output frame with frame_num #2, poc 8 (dpb. dpb.size=16,
dpb.used_size=1) 
	duplicate frame_num in short-term reference picture buffer 
	and the decoder stops. I am not sure why this happens. Can
anyone please explain the reason and what can be done in this regard. It
is really urgent for me and so I request you to kindly help me out with
your suggestions.....
	-- 
	Thanks and Regards
	Mohammad Khalil Issa Jubran
	State University Of New York at Buffalo "SUNY at Buffalo"
	School of Electrical Engineering         
	PhD. Student.
	Tel. 001-716-400-4677
	Mailing Address
	         123 Hawthorne ave "upper"
	         Tonawanda - NY - 14223 , USA
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051114/3e13f618/attachment.html
From garysull  windows.microsoft.com  Mon Nov 14 13:24:07 2005
From: garysull  windows.microsoft.com (Gary Sullivan)
Date: Tue Nov 15 08:32:27 2005
Subject: [Mp4-tech] about init reference list
Message-ID: <91D7F2CEE3425A4A9D11311D09FCE246118627DD@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
If both lists are the same, accessing list0 and list1 amounts to having
duplicate syntax to signal the same thing in many cases (e.g., when not
using explicit weighted prediction).  That's ordinarily bad from a
compression point of view.
Perhaps we should have been even more agressive with this rule, such as
swapping out the first entry whenever just that one entry is the same in
both lists, instead of requiring both entire lists to be the same in
order to initiate the swap.  But in any case it doesn't take very many
bits for the encoder to re-order the lists if something other than the
default result is desired.
Best Regards,
Gary Sullivan
________________________________
	From: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of China264
	Sent: Sunday, November 13, 2023 9:24 PM
	To: mp4-tech@lists.mpegif.org
	Subject: [Mp4-tech] about init reference list
	Dear all,
	               In "   Initialisation process for reference
picture lists for B slices in fields", there are a sentence as
following, 
	When the reference picture list RefPicList1 has more than one
entry and it is identical to the reference picture list RefPicList0, the
first two entries RefPicList1[0] and RefPicList1[1] are switched. 
	I have checked the source code of JM 86. The source code do it
like this.
	Can some expert tell me why do like this?
	thanks!
	jeff
	-- 
	-----------------------------------------------
	Video chip design BBS
	China264.68L.com
	China264@gmail.com 
	------------------------------------------------ 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051114/ecc9e85e/attachment.html
From garysull  windows.microsoft.com  Mon Nov 14 14:48:23 2005
From: garysull  windows.microsoft.com (Gary Sullivan)
Date: Tue Nov 15 08:32:35 2005
Subject: [Mp4-tech] H264 Deblocking
Message-ID: <91D7F2CEE3425A4A9D11311D09FCE24611862975@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
Giridhar,
There are a lot of details to keep track of in the deblocking filter
process.  I suggest reading the details of the specification carefully.
Alpha and Beta are derived from IndexA and IndexB, which depend on qPav.
qPav is the average of the QP on the two sides of an edge to be
filtered.  For the interior edges of a macroblock, QP will be the same
on both sides of the edge.  However, for edges that form a boundary
between two different macroblocks, the QP may be different on each side
of the edge.
When MbaffFrameFlag is equal to 0, there are two such boundary cases --
for the left neighbor MB and for the top neighbor MB.  So when
MbaffFrameFlag is 0, there will be a total of three distinct values of
alpha and beta for luma, and three for chroma, per macroblock.
When MbaffFrameFlag is equal to 1, I believe there can be up to five
distinct values, since there can be two left-neighbor macroblocks and
two top-neighbor macroblocks.
This analysis doesn't count the filtering of the right and bottom edges
of the macroblock, since those edges will be considered the left and top
edges of some other macroblock (or will be the edges of the picture and
thus will not get filtered).
Best Regards,
Gary Sullivan
________________________________
	From: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Giridhar
	Sent: Monday, November 14, 2023 4:47 AM
	To: mp4-tech@lists.mpegif.org
	Subject: [Mp4-tech] H264 Deblocking
	Hello all,
	My question is filter decision, as deblocking is done for 4x4
block boundaries in a MacroBlock do the values of alpha and beta vary
for each 4x4 block boundary or will it be constant for entire MB?
	TIA
	giridhar
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051114/470a3325/attachment-0001.html
From kj090177  yahoo.com  Mon Nov 14 15:23:41 2005
From: kj090177  yahoo.com (KJ Yang)
Date: Tue Nov 15 08:32:40 2005
Subject: [Mp4-tech] Video codec benchmarking
In-Reply-To: <20051111121315.70547.qmail@web52810.mail.yahoo.com>
Message-ID: <20051114232341.74252.qmail@web53510.mail.yahoo.com>
Hi! all:
Does anyone know any tool or document for video codec
benchmarking. thanks a lot
-K.J.
__________________________________ 
Yahoo! Mail - PC Magazine Editors' Choice 2005 
http://mail.yahoo.com
From skb3  buffalo.edu  Mon Nov 14 19:51:18 2005
From: skb3  buffalo.edu (Saurav Bandyopadhyay)
Date: Tue Nov 15 08:32:46 2005
Subject: [Mp4-tech] problem with H.263 codec version JM10.1
	"duplicateframe_num in short-term reference picture buffer "
In-Reply-To: <91D7F2CEE3425A4A9D11311D09FCE246118627D7@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
Message-ID: <200511150051.jAF0pMCV010391@lists1.magma.ca>
Dear Dr. Sullivan,
Thank you for referring to me. I am replying below to Mohammad's question:
Mohammad et al.:  From the error message you are getting, I think you are
using IDR frames and a bit stream with non reference frames. Please set the
POC gap and reference POC gap accordingly in the decoder configuration file
i.e. reference POC gap = 2 if you are using IPP and 4 if using IbP / IpP.
Also, you need to set POC gap to 2 if using IPP /IbP/IpP.
JM 10.1 gives you an option to conceal an entire frame loss using frame copy
or motion copy concealment algorithms.  This scheme was proposed to conceal
an entirely lost frame in a compressed video bit stream. The proposed scheme
targets at low bit rate video transmission applications using H.264/AVC. In
case you want to enable these options set the decoder file error concealment
entry to 1: Frame Copy or 2: Motion Copy. Setting it to "0" will disable the
frame loss concealment option.  However, it will still be able to conceal
slice losses (perhaps this is your case). Please set the configuration file
entries accordingly.
Hope this helps.
Best regards,
Saurav
----------------------------------------------------------------------
Saurav K. Bandyopadhyay
Ph.D Candidate & Teaching Assistant
?????????????????????????????
Multimedia Communications Lab???
Dept of Electrical Engineering
Univ at Buffalo, The State Univ of NY???
332 Bonner Hall,????????????????????????
Buffalo, NY - 14260?????????????????????
USA????????????????????????????????????
?????????????????????????????????????
Tel: +1 (716) 830 6271???????????????
Web: http://www.buffalo.edu/~skb3
______________________________________________________________________
________________________________________
From: Gary Sullivan [mailto:garysull@windows.microsoft.com] 
Sent: Monday, November 14, 2023 4:23 PM
To: Mohammad Jubran; mp4-tech@lists.mpegif.org
Cc: skb3@buffalo.edu
Subject: RE: [Mp4-tech] problem with H.263 codec version JM10.1
"duplicateframe_num in short-term reference picture buffer "
Mohammad et al,
?
Probably one of the best people to talk to about this is your SUNY-Buffalo
colleague Saurav Bandyopadhyay [skb3@buffalo.edu].
?
Graceful behavior of a decoder under packet loss conditions is a desirable,
but not mandatory, feature for implementations (from the standardization
perspective, that is -- perhaps not from the marketability perspective).
?
Best Regards,
?
Gary Sullivan
________________________________________
From: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Mohammad Jubran
Sent: Saturday, November 12, 2023 6:24 PM
To: mp4-tech@lists.mpegif.org; eng.mjubran@gmail.com
Subject: [Mp4-tech] problem with H.263 codec version JM10.1
"duplicateframe_num in short-term reference picture buffer "
Hello
?
I am using H.264 codec version JM?10.1 and I have encoded my video in the
RTP packet format .DP is enabled.?I use some packet dropping scheme to drop
the RTP packets (packet is a partition?of a slice)?as per a given Packet
Loss Rate and feed the rest to the decoder. But for any case, I get an error
which says
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=4 
output frame with frame_num #0, poc 0 (dpb. dpb.size=16, dpb.used_size=4 ) 
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=3 
output frame with frame_num #2, poc 2 (dpb. dpb.size=16, dpb.used_size=3) 
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=2 
output frame with frame_num #1, poc 4 (dpb. dpb.size=16, dpb.used_size=2) 
Flush remaining frames from dpb. dpb.size=16, dpb.used_size=1 
output frame with frame_num #2, poc 8 (dpb. dpb.size=16, dpb.used_size=1) 
duplicate frame_num in short-term reference picture buffer 
and the decoder stops. I am not sure why this happens. Can anyone please
explain the reason and what can be done in this regard. It is really urgent
for me and so I request you to kindly help me out with your suggestions.....
-- 
Thanks and Regards
Mohammad Khalil Issa Jubran
State University Of New York at Buffalo "SUNY at Buffalo"
School of Electrical Engineering???????? 
PhD. Student.
Tel. 001-716-400-4677
Mailing Address
???????? 123 Hawthorne ave "upper"
???????? Tonawanda - NY - 14223 , USA
?
From min-mxa  cnt.necst.nec.co.jp  Tue Nov 15 09:57:58 2005
From: min-mxa  cnt.necst.nec.co.jp (min-mxa@cnt.necst.nec.co.jp)
Date: Tue Nov 15 08:32:52 2005
Subject: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS
Message-ID: <20051115095758min-mxa@mail.jp.nec.com>
Hi experts!
I am trying to multiplex a MPEG4 ASP video ES into a MPEG2 TS.
As you know B VOPs are allowed in MPEG4 ASP.
The MPEG4 ASP video ES is generated by a encoder board and 
when each AU is being outputed the responding PTS and DTS can be geted.
So when a I or P VOP is packed into PES packet,how shall I set the DTS 
in the PESHeader?
thank you very much!
From sh  upcsurpass.com  Tue Nov 15 09:15:58 2005
From: sh  upcsurpass.com (Song)
Date: Tue Nov 15 08:32:59 2005
Subject: [Mp4-tech] H264 please help me about the transform algorithm?
Message-ID: <001d01c5e982$2ee5f590$6b01a8c0@hongsong>
Dear Experts,
I am reading the decode parts of JM10.1 source code, but I cannot understand the fast algorithm of I-DCT transform. Does anyone give me some links to find some background knowledge about it?
Regards.
Bill
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051115/6a4d0eb5/attachment.html
From china264  gmail.com  Tue Nov 15 10:33:21 2005
From: china264  gmail.com (China264)
Date: Tue Nov 15 08:33:05 2005
Subject: [Mp4-tech] about init reference list
In-Reply-To: <91D7F2CEE3425A4A9D11311D09FCE246118627DD@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
References: <91D7F2CEE3425A4A9D11311D09FCE246118627DD@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
Message-ID: <ac4cd46a0511141733w5d0d7ce4la24d7da3d0a714bc@mail.gmail.com>
Thanks for your answer!
 But I am also wondering about this special case.
 From the source code of JM86, the list0 is ordered as following,
 The ascended order short term reference frame which poc is less than
current image poc, the descended order short term reference frame which poc
is more than current image poc, and the long term reference frame.
 The list1 is
  The descended order short term reference frame which poc is more than
current image poc, the ascended order short term reference frame which poc
is less than current image poc, and the long term reference frame.
 Only the short term part of list 0 and list1 have been exchanged. The long
term reference frame is same.
 So if the list 0 and list 1 is same, there are only two case.
 1 only forward or backword short term reference image is in list0 and
list1. If the forward and backword reference image are both in list0 and
list1, the backward and forward reference frame should be same! But it's
impossible!
 2 There are no any short term reference image in list0 and list1. only long
term reference image is in list.
 The above thinking is right or not?
 Thanks for your help again!!!
 On 11/15/05, Gary Sullivan <garysull@windows.microsoft.com> wrote:
>
> If both lists are the same, accessing list0 and list1 amounts to having
> duplicate syntax to signal the same thing in many cases (e.g., when not
> using explicit weighted prediction). That's ordinarily bad from a
> compression point of view.
>  Perhaps we should have been even more agressive with this rule, such as
> swapping out the first entry whenever just that one entry is the same in
> both lists, instead of requiring both entire lists to be the same in order
> to initiate the swap. But in any case it doesn't take very many bits for the
> encoder to re-order the lists if something other than the default result is
> desired.
>  Best Regards,
>  Gary Sullivan
>
>  ------------------------------
> *From:* mp4-tech-bounces@lists.mpegif.org [mailto:
> mp4-tech-bounces@lists.mpegif.org] *On Behalf Of *China264
> *Sent:* Sunday, November 13, 2023 9:24 PM
> *To:* mp4-tech@lists.mpegif.org
> *Subject:* [Mp4-tech] about init reference list
>
>  Dear all,
>  In " Initialisation process for reference picture lists for B slices in
> fields", there are a sentence as following,
>  When the reference picture list RefPicList1 has more than one entry and
> it is identical to the reference picture list RefPicList0, the first two
> entries RefPicList1[0] and RefPicList1[1] are switched.
>
> I have checked the source code of JM 86. The source code do it like this.
>  Can some expert tell me why do like this?
>  thanks!
>  jeff
>
>
> --
> -----------------------------------------------
> Video chip design BBS
> China264.68L.com <http://china264.68l.com/>
> China264@gmail.com
> ------------------------------------------------
>
>
--
-----------------------------------------------
Video chip design BBS
China264.68L.com <http://China264.68L.com>
China264@gmail.com
------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051115/46a70467/attachment.html
From prashanthp  gmail.com  Tue Nov 15 09:42:06 2005
From: prashanthp  gmail.com (Prashanth P)
Date: Tue Nov 15 08:33:11 2005
Subject: [Mp4-tech] Re: Mp4-tech Digest, Vol 28, Issue 9
In-Reply-To: <200511141414.jAEEAOLU023641@lists1.magma.ca>
References: <200511141414.jAEEAOLU023641@lists1.magma.ca>
Message-ID: <71c3c4ac0511142012y7f2e052ei54e90b6088b5126a@mail.gmail.com>
Hi,
To extract the Audio and Video content from the MP4 File you can use the
Virtual Dub tool. you can download the tool from the net, which is a
freeware.
 -Prashanth
 On 11/14/05, mp4-tech-request@lists.mpegif.org <
mp4-tech-request@lists.mpegif.org> wrote:
>
> Send Mp4-tech mailing list submissions to
> mp4-tech@lists.mpegif.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://lists.mpegif.org/mailman/listinfo/mp4-tech
> or, via email, send a message with subject or body 'help' to
> mp4-tech-request@lists.mpegif.org
>
> You can reach the person managing the list at
> mp4-tech-owner@lists.mpegif.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Mp4-tech digest..."
>
>
> Today's Topics:
>
> 1. Re: .mp4 file format (Dave Singer)
> 2. RE: Iprof --Complexity analysis (Tourapis, Alexis)
> 3. duplicate frame_num in short-term reference picture buffer
> (Mohammad Jubran)
> 4. problem with H.263 codec version JM10.1 "duplicate frame_num
> in short-term reference picture buffer " (Mohammad Jubran)
> 5. about init reference list (China264)
> 6. Re: Some doubts in video (Herbert Thoma)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 11 Nov 2023 10:23:49 -0800
> From: Dave Singer <singer@apple.com>
> Subject: Re: [Mp4-tech] .mp4 file format
> To: Smikky Sahaya Singh <smikkyssingh@gmail.com>,
> mp4-tech@lists.mpegif.org
> Message-ID: <p06230912bf9a91f572fe@[17.219.157.83 <http://17.219.157.83>]>
> Content-Type: text/plain; charset="us-ascii"
>
> At 16:59 +0530 11/11/05, Smikky Sahaya Singh wrote:
> >Hi,
> >
> >How to extract the video and audio part from the .mp4 container?
> >
> >Is there any document which explains the .mp4 file format(Explaining
> >the Packing of Video and Audio)?
>
> ISO/IEC 14496-14 (mpeg-4 specific handling), ISO/IEC 14496-12 (all
> the structure)
> part 12 is freely available from ISO
>
> you might also look at
> ISMA 1.0 and ISMA 2.0, which are also free.
>
> >
> >Thanks in Advance.
> >JC
> >
> >
> >_______________________________________________
> >NOTE: Please use clear subject lines for your posts. Include [audio,
> >[video], [systems], [general] or another apppropriate identifier to
> >indicate the type of question you have.
> >
> >Note: Conduct on the mailing list is subject to the Antitrust
> >guidelines found at
> >http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
>
>
> --
> David Singer
> Apple Computer/QuickTime
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: /pipermail/mp4-tech/attachments/20051111/ccfb3c7a/attachment-
> 0001.html
>
> ------------------------------
>
> Message: 2
> Date: Fri, 11 Nov 2023 12:17:41 -0800
> From: "Tourapis, Alexis" <alexis.tourapis@dolby.com>
> Subject: RE: [Mp4-tech] Iprof --Complexity analysis
> To: "mayank agarwal" <may_ank77@yahoo.com>, "Humaira"
> <humaira.nisar@gmail.com>, <mp4-tech@lists.mpegif.org>
> Message-ID:
> <7272EE229DA1AA48B47EBDC47EB0C68B019D9167@sapphire.dolby.net>
> Content-Type: text/plain; charset="utf-8"
>
> Dear Humaira and Mayank,
>
> I am not sure if Iprof is still available or supported anymore. Last time
> that I had checked the official website (i.e.
> http://www.lis.ei.tum.de/research/bv/topics/method/e_iprof.html ) was
> providing a dead link.
> However, as many people would tell you, you may instead use many free or
> commercial profiler applications. You may wish to check for example gprof,
> vtune, codeanalys, the devpartner profiler. Note that there is also another
> profiler named iprof at :
>
> http://silverspaceship.com/src/iprof/
>
> but this seems to have no relationship with the original.
>
> You may also want to check the atomium suite at
> http://www.imec.be/design/atomium/toolsuite.shtml <
> http://www.imec.be/design/atomium/toolsuite.shtml> while, some nice work
> (and obvious quite relevant) has also been done from EPFL. I would suggest
> getting hold of their MPEG M9553 contribution, named "Complexity Evaluation
> of Different Configurations of the JVT Codec". In fact you may download
> their full contribution also from the JVT site (see
> http://ftp3.itu.ch/av-arch/jvt-site/2003_03_Pattaya/, contribution
> JVT-G038, which also includes their software).
>
> With regards to some of the questions on motion estimation, motion
> estimation does not need to be only part of an encoding solution. Motion
> estimation is in fact used in many different systems for various purposes
> ranging from pre and post filtering, segmentation, registration, scene
> analysis etc. Therefore, the answer is yes you can implement motion
> estimation as a standalone module. However (and this is a big however), one
> thing you need to consider is the fact that within an encoder you actually
> introduce artifacts due to the encoding process itself which would have an
> immediate impact on the ME process as well (assuming that you consider
> reconstructed images as references). Note also that a video codec, such as
> H.264, contains also several other components that would impact motion
> estimation, such as Rate Distortion based optimized mode decision, Rate
> control etc which will also affect your decisions and your final motion
> field. Many algorithms also at times consider mot!
> ion fields of previously coded pictures to reduce complexity which would
> require you a bit more work to add in an outside tool (i.e. you would need
> to create a buffer as well).
>
> Your second question is maybe a bit tough to answer considering that the
> answer depends on many parameters. I.e. is your implementation a hardware
> or software one, what constraints do you have in terms of memory and
> processing power, can you use parallelization instructions or not, do you
> have issues with branching etc.
>
> You may wish to check JVT-O079 for information on some of the schemes the
> JM is using. Note though that this document is currently being updated and
> hopefully there will be a much better version before the next meeting
> describing more accurately the software's behavior and the algorithms
> available in the software. Note that we are also trying to rearchitecture
> somewhat the JM encoder and decoder as to make them faster and a bit more
> friendlier.
>
> Best regards,
>
> Alexis
>
>
> -----Original Message-----
> From: mp4-tech-bounces@lists.mpegif.org on behalf of mayank agarwal
> Sent: Thu 11/10/2023 9:04 PM
> To: Humaira; mp4-tech@lists.mpegif.org
> Cc:
> Subject: Re: [Mp4-tech] Iprof --Complexity analysis
>
>
>
> Please forward the reply of this mail to me also.
> I also want that.Also if some one can clarify the
> following things then it will be of great help.
>
> 1.Is it possible to implement motion estimation can be
> implemented as standalone application upon any image
> which we can see in YUV viewer and then analyse the
> results and based upon that we can deduce that if it
> is made a part of any algorithm like H.264/MPEG4 then
> what will be the performance in terms of frame/s and
> bit rate.
>
> 2.How do we judge the suitability of particular motion
> estimation algorithm.
>
> 3.In the JM reference code for H.264 a combination of
> integer pixel motion estimation and fractional pixel
> estimation is used and rate distortion optimization
> is also used.My questins are:
> 1.If there are any documents which explain in
> detail how to do the above things.
> 2.If any real time implementation of H.264 supports
> that kind of technique and the performance results
> from that.
>
>
>
> --- Humaira <humaira.nisar@gmail.com> wrote:
>
> > Hi
> > Do anyone has an idea how to get iprof. Or is there
> > some other good
> > software that helps in the complexity analysis of
> > motion estimation
> > algorithms.
> > Also i would like to know about some place that will
> > help in the programming
> > of motion estimation algorithms
> > Thanks
> > > _______________________________________________
> > NOTE: Please use clear subject lines for your posts.
> > Include [audio, [video], [systems], [general] or
> > another apppropriate identifier to indicate the type
> > of question you have.
> >
> > Note: Conduct on the mailing list is subject to the
> > Antitrust guidelines found at
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
>
>
>
>
>
> __________________________________
> Yahoo! Mail - PC Magazine Editors' Choice 2005
> http://mail.yahoo.com
> _______________________________________________
> NOTE: Please use clear subject lines for your posts. Include [audio,
> [video], [systems], [general] or another apppropriate identifier to indicate
> the type of question you have.
>
> Note: Conduct on the mailing list is subject to the Antitrust guidelines
> found at
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
>
>
>
>
> -----------------------------------------
> This message (including any attachments) may contain confidential
> information intended for a specific individual and purpose. If you are
> not the intended recipient, delete this message. If you are not the
> intended recipient, disclosing, copying, distributing, or taking any
> action based on this message is strictly prohibited.
>
>
> ------------------------------
>
> Message: 3
> Date: Sat, 12 Nov 2023 21:10:37 -0500
> From: Mohammad Jubran <eng.mjubran@gmail.com>
> Subject: [Mp4-tech] duplicate frame_num in short-term reference
> picture buffer
> To: mp4-tech@lists.mpegif.org
> Message-ID:
> <acb31d960511121810s351506dbt44d9d1883992805e@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hello
> I am trying to test the optimality of H264 in wireless systesm. Droping
> some partitions give the following message while decoding :
> Flush remaining frames from dpb. dpb.size=16, dpb.used_size=4
> output frame with frame_num #0, poc 0 (dpb. dpb.size=16, dpb.used_size=4)
> Flush remaining frames from dpb. dpb.size=16, dpb.used_size=3
> output frame with frame_num #2, poc 2 (dpb. dpb.size=16, dpb.used_size=3)
> Flush remaining frames from dpb. dpb.size=16, dpb.used_size=2
> output frame with frame_num #1, poc 4 (dpb. dpb.size=16, dpb.used_size=2)
> Flush remaining frames from dpb. dpb.size=16, dpb.used_size=1
> output frame with frame_num #2, poc 8 (dpb. dpb.size=16, dpb.used_size=1)
> duplicate frame_num in short-term reference picture buffer
> Can Anyone help me figure out why is that happening?
>
> --
> Regards
>
> Mohammad Khalil Issa Jubran
> State University Of New York at Buffalo "SUNY at Buffalo"
> School of Electrical Engineering
> PhD. Student.
> Tel. 001-716-400-4677
> Mailing Address
> 123 Hawthorne ave "upper"
> Tonawanda - NY - 14223 , USA
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: /pipermail/mp4-tech/attachments/20051112/a6de3d43/attachment-
> 0001.html
>
> ------------------------------
>
> Message: 4
> Date: Sat, 12 Nov 2023 21:23:31 -0500
> From: Mohammad Jubran <eng.mjubran@gmail.com>
> Subject: [Mp4-tech] problem with H.263 codec version JM10.1 "duplicate
> frame_num in short-term reference picture buffer "
> To: mp4-tech@lists.mpegif.org, eng.mjubran@gmail.com
> Message-ID:
> <acb31d960511121823y3790aba3l1c22baac078b7884@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hello
> *I am using H.264 codec version JM 10.1 and I have encoded my **video in
> the RTP packet format .DP is enabled. I use some packet dropping scheme to
> *
> *drop the RTP packets (packet is a partition of a slice) as per a given
> Packet Loss Rate and feed the rest **to the decoder. But for any case, I
> get
> an error which says
> **
> **Flush remaining frames from dpb. dpb.size=16, dpb.used_size=4
> output frame with frame_num #0, poc 0 (dpb. dpb.size=16, dpb.used_size=4)
> Flush remaining frames from dpb. dpb.size=16, dpb.used_size=3
> output frame with frame_num #2, poc 2 (dpb. dpb.size=16, dpb.used_size=3)
> Flush remaining frames from dpb. dpb.size=16, dpb.used_size=2
> output frame with frame_num #1, poc 4 (dpb. dpb.size=16, dpb.used_size=2)
> Flush remaining frames from dpb. dpb.size=16, dpb.used_size=1
> output frame with frame_num #2, poc 8 (dpb. dpb.size=16, dpb.used_size=1)
> duplicate frame_num in short-term reference picture buffer
> **
> **and the decoder stops. I am not sure why this happens. Can anyone
> **please
> explain the reason and what can be done in this regard. It is **really
> urgent for me and so I request you to kindly help me out with **your
> suggestions.....
> *
>
> --
> Thanks and Regards
>
> Mohammad Khalil Issa Jubran
> State University Of New York at Buffalo "SUNY at Buffalo"
> School of Electrical Engineering
> PhD. Student.
> Tel. 001-716-400-4677
> Mailing Address
> 123 Hawthorne ave "upper"
> Tonawanda - NY - 14223 , USA
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: /pipermail/mp4-tech/attachments/20051112/4141e459/attachment-
> 0001.html
>
> ------------------------------
>
> Message: 5
> Date: Mon, 14 Nov 2023 14:24:03 +0900
> From: China264 <china264@gmail.com>
> Subject: [Mp4-tech] about init reference list
> To: mp4-tech@lists.mpegif.org
> Message-ID:
> <ac4cd46a0511132124o6955f1dqd33bfae2a809dae5@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Dear all,
> In " Initialisation process for reference picture lists for B slices in
> fields", there are a sentence as following,
> When the reference picture list RefPicList1 has more than one entry and it
> is identical to the reference picture list RefPicList0, the first two
> entries RefPicList1[0] and RefPicList1[1] are switched.
>
> I have checked the source code of JM 86. The source code do it like this.
> Can some expert tell me why do like this?
> thanks!
> jeff
>
>
> --
> -----------------------------------------------
> Video chip design BBS
> China264.68L.com <http://China264.68L.com> <http://China264.68L.com>
> China264@gmail.com
> ------------------------------------------------
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: /pipermail/mp4-tech/attachments/20051114/9ef6db9f/attachment-
> 0001.html
>
> ------------------------------
>
> Message: 6
> Date: Mon, 14 Nov 2023 12:31:24 +0100
> From: Herbert Thoma <tma@iis.fhg.de>
> Subject: Re: [Mp4-tech] Some doubts in video
> To: mayank agarwal <may_ank77@yahoo.com>
> Cc: mp4-tech@lists.mpegif.org, c6x@yahoogroups.com
> Message-ID: <4378758C.8070809@iis.fhg.de>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> mayank agarwal wrote:
> > Hi all,
> > Recently some person asked me some questions about
> > video.I found myself cluless on those questions.
> > Please help me to clarify on the following topics.
> >
> > 1.In MPEG-4 simple profile how many motion vectors are
> > transmitted from encoder to decoder or generalizing
> > the above can anybody tell how to determine the number
> > of motion vectors being transmitted from encoder to
> > decoder.
>
> Hmm, decode the bitstream? The extreme cases are:
> - no motion vectors at all (e.g. IVOPs)
> - all macroblocks coded with 4 vectors (this is
> probably not very likely)
>
> There is a mechanism called "complexity estimation"
> in the standard but as far as I know nobody uses this ...
>
> > 2.Suppose in the video encoding we don't do
> > quantization then what will be affect on final encoded
> > stream.
>
> That's not allowed by the standard. So your stream
> will not be valid.
>
> > 3.Suppose we are encoding QCIF sized video then what
> > will be the minimum size of decoded buffer.I suppose
> > it is 176*144+2*176/2*144/2.Please correct me if i am
> > wrong.
>
> That's correct.
>
> > 4.Suppose we are doing some video integration and we
> > found out that DSP memory is not enough for the
> > encoded stream then how can we increase DSP memory.
>
> I don't understand this question. Add more memory to
> the DSP?
>
> Kind regards,
> Herbert.
>
> > Thanks and Regards,
> > Mayank
> >
> >
> >
> > __________________________________
> > Yahoo! FareChase: Search multiple travel sites in one click.
> > http://farechase.yahoo.com
> > _______________________________________________
> > NOTE: Please use clear subject lines for your posts. Include [audio,
> [video], [systems], [general] or another apppropriate identifier to indicate
> the type of question you have.
> >
> > Note: Conduct on the mailing list is subject to the Antitrust guidelines
> found at
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
> >
>
> --
> Herbert Thoma
> Head of Video Group
> Multimedia Realtime Systems Department
> Fraunhofer IIS
> Am Wolfsmantel 33, 91058 Erlangen, Germany
> Phone: +49-9131-776-323
> Fax: +49-9131-776-399
> email: tma@iis.fhg.de
> www: http://www.iis.fhg.de/
>
> ------------------------------
>
> _______________________________________________
> Please use clear subject lines for your posts. Include [audio, [video],
> [systems], [general] or another apppropriate identifier to indicate the type
> of question you have.
>
> Conduct on the mailing list is subject to the Antitrust guidelines found
> at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
>
> End of Mp4-tech Digest, Vol 28, Issue 9
> ***************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051115/2d54ca59/attachment-0001.html
From giri_tammana  yahoo.co.in  Tue Nov 15 17:35:36 2005
From: giri_tammana  yahoo.co.in (Giridhar)
Date: Tue Nov 15 14:21:42 2005
Subject: [Mp4-tech] H264 deblocking
Message-ID: <!~!UENERkVCMDkAAQACAAAAAAAAAAAAAAAAABgAAAAAAAAA7QqDKrLZ/keAzZXjHS06VMKAAAAQAAAAeOt3NgjR5UuBuJ2YfznfeAEAAAAA@yahoo.co.in>
Thank you for your reply Gary. 
Still I have one doubt regarding the calculation of  strength in reference
software. In loopfilter.c Getstrength is called for calculating 4 4x4 block
edges with strength[16].
In Getstrength it calculates 16 values of strength. If we are calculating
strength for one vertical stripe then why 16 values are calculated instead
of 4 values for 4 blocks edges in one vertical stripe?
Correct me if I understood it wrong.
TIA
Giridhar
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051115/511af35a/attachment.html
From mykerekes  eml.cc  Tue Nov 15 05:52:10 2005
From: mykerekes  eml.cc (hungary)
Date: Tue Nov 15 14:30:17 2005
Subject: [Mp4-tech] CABAC versus adaptive huffmann
In-Reply-To: <71c3c4ac0511142012y7f2e052ei54e90b6088b5126a@mail.gmail.com>
References: <200511141414.jAEEAOLU023641@lists1.magma.ca>
	<71c3c4ac0511142012y7f2e052ei54e90b6088b5126a@mail.gmail.com>
Message-ID: <1132062730.17268.247550154@webmail.messagingengine.com>
Hai experts,
can any one put some light on 
how CABAC differs from adaptive huffmann coding?
Or both of them are same ?
-- 
  hungary
  mykerekes@eml.cc
-- 
http://www.fastmail.fm - The way an email service should be
From fful  conncoll.edu  Tue Nov 15 08:59:59 2005
From: fful  conncoll.edu (Frank Fulchiero)
Date: Wed Nov 16 14:47:15 2005
Subject: [Mp4-tech] Re: Audio /Video   Podcast
In-Reply-To: <200511151158.jAFBweK6008956@lists1.magma.ca>
References: <200511151158.jAFBweK6008956@lists1.magma.ca>
Message-ID: <FAC10C77-47AB-43B6-BD99-C47DF3659BCA@conncoll.edu>
> From: Joseph Lahart <lahmurmusic@mac.com>
> Subject: [Mp4-tech] Audio /Video   Podcast.
>
> I have .mac account and want to up load this to a pod cast. I have red
> some articals and gone to various web sites but
> am no closer. There is plenty of info about audio but i have audio and
> video. Any help would be appreciated.
Look at http://www.vodcast.nl/howto.html
and http://www.apple.com/quicktime/tutorials/videopodcasts.html
I wrote a simple tutorial for audio only at
http://video.conncoll.edu/podcast/index.html
There is no major difference between audio and video, I believe.
For example, in iTunes, subscribe to
http://sethpowsner.home.comcast.net/test.xml
If you look at the code, you can figure it out.
You might be better off asking for more help at podcasting-related  
sites, like the forum at Podcasting News.
http://www.podcastingnews.com/
Good luck,
Frank Fulchiero
Digital Media Specialist
Connecticut College
From rlei  ati.com  Tue Nov 15 12:16:08 2005
From: rlei  ati.com (Ryan Lei)
Date: Wed Nov 16 14:47:21 2005
Subject: [Mp4-tech] [System][AAC][3gp]: SBR suooprt in .3gp file format
Message-ID: <C6D6960055FE3E4FBE5D4A8D40D8F0760133E284@torcaexmb2.atitech.com>
Hi,
Is there any document available about how to support AAC SBR in 3gp file
format? What is the difference in the 'mp4a' box for SBR and AAC
baseline?
Thanks
Ryan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051115/91ba861c/attachment.html
From Yubin.Zhu  esstech.com  Tue Nov 15 09:29:02 2005
From: Yubin.Zhu  esstech.com (Yu-Bin Zhu)
Date: Wed Nov 16 14:47:29 2005
Subject: [Mp4-tech] Video codec benchmarking
Message-ID: <89976087EE37FA4D8684C71F318DD7D9026D363D@ess2kmail.essnet.com>
You may check with BDTI(www.bdti.com). They might have benchmarks for video codecs.
-----Original Message-----
From: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org]On Behalf Of KJ Yang
Sent: Monday, November 14, 2023 3:24 PM
To: mp4-tech@lists.mpegif.org
Subject: [Mp4-tech] Video codec benchmarking
Hi! all:
Does anyone know any tool or document for video codec
benchmarking. thanks a lot
-K.J.
__________________________________ 
Yahoo! Mail - PC Magazine Editors' Choice 2005 
http://mail.yahoo.com
_______________________________________________
NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have.
Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
From john.l.smith  titan.com  Tue Nov 15 13:58:19 2005
From: john.l.smith  titan.com (John Smith)
Date: Wed Nov 16 14:47:34 2005
Subject: [Mp4-tech] Comparison of H.264 coding efficiency of main vs.
	baseline profile?
Message-ID: <9664A65FCDEDAF498FB654F8425D88D10A1D4B@mail.advprod.net>
I've looked around a bit, but so far have not found any papers comparing
the two profiles against each other.
Can anyone here suggest how much does main improve baseline for the same
PSNR?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051115/ed1f98d4/attachment.html
From sh  upcsurpass.com  Wed Nov 16 09:57:15 2005
From: sh  upcsurpass.com (Song)
Date: Wed Nov 16 14:47:39 2005
Subject: [Mp4-tech]some puzzles about H264 stream.
Message-ID: <003601c5ea51$16dd9c80$6b01a8c0@hongsong>
Dear Experts,
I am new about H264, I am wondering that how to find the information such as Aspect Ratio, Frame Rate from a H.264 elementary stream? I didn't find the related syntax elements. 
The second question is how to find the Pictures end? In H.264, slice structure is the top level syntax about a picture, how to know the slice is the new starting of next picture?
Are there any identifier of a picture? or What syntax element can be used as picture ID? that is to say, the element or some number can denote a picture uniquely?
Thank you in advance,
Regards.
Bill
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051116/6abef794/attachment.html
From luokai  zju.edu.cn  Wed Nov 16 14:33:01 2005
From: luokai  zju.edu.cn (luokai)
Date: Wed Nov 16 14:47:45 2005
Subject: [Mp4-tech]Assign a LongTermFrameIdx to a short-term ref pic 
Message-ID: <332122990.18575@eyou.net>
Hello, expert:
When assigning a LongTermFrameIdx to a short-term ref pic
it is said ". field_pic_flag is equal to 1, the marking of the short-term
ref field specified by
picNumX is changed from 'used for short-term ref' to 'used for long-term
ref' ."
My doubt is can this situation happen: after this operation, top field of a
frame is "used for 
Short-term ref" and bottom field of the same frame is "used for long-term
ref".
Thanks!
luokai
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051116/19ea4440/attachment.html
From tom_aac  yahoo.com  Wed Nov 16 00:54:27 2005
From: tom_aac  yahoo.com (Tom Colls)
Date: Wed Nov 16 14:47:50 2005
Subject: [Mp4-tech] [AAC] LTP for short windows
Message-ID: <20051116085427.46258.qmail@web36415.mail.mud.yahoo.com>
Hi experts,
  I was little confused about the way LTP is done for short windows(for AAC decoder). The things that have been bugging me include
  1. how ltp_short_lag is combined with ltp_lag ? The standard doesnt say anything about this , however the reference software does throw some light on this. That is, it it has some DPCM kind of encoding(like used for spectral data). Is it supposed to be handled the way its done in the reference software ?(http://www.mp3-tech.org/programmer/sources/refsoft990809.tgz)
  2. how is ltp done for short windows ?
  hows the total lag (computed as above) used for short windows. Is the MDCT performed in the LTP loop a 256 point MDCT , or do we still use a 2048 point MDCT over the whole frame(current frame + past frame) once ? , like the way its done for long windows.
  Any help would be very much appreciated,
  Regards
---------------------------------
 Yahoo! FareChase - Search multiple travel sites in one click.  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051116/a285f87c/attachment.html
From vishu_dash  yahoo.com  Wed Nov 16 11:18:04 2005
From: vishu_dash  yahoo.com (Deshpande,Vishvanath)
Date: Wed Nov 16 14:47:55 2005
Subject: [Mp4-tech] Regarding conformance testing of HE-AAC
In-Reply-To: <OF81ECB9F8.F2F9D761-ONC12570B4.0001D470-412570B4.0003F880@codingtechnologies.com>
Message-ID: <20051116111804.32795.qmail@web60717.mail.yahoo.com>
Hi,
Can any one pls tell me where do we get 5.1 ch
conformance streams for AAC decoder(AAC-Lc and HE
AAC)?
Thanks in advance,
regards,
-Vishvanath
___________________________________________________________ 
How much free photo storage do you get? Store your holiday 
snaps for FREE with Yahoo! Photos http://uk.photos.yahoo.com
From garysull  windows.microsoft.com  Wed Nov 16 10:57:29 2005
From: garysull  windows.microsoft.com (Gary Sullivan)
Date: Thu Nov 17 13:37:01 2005
Subject: [Mp4-tech]some puzzles about H264 stream.
Message-ID: <91D7F2CEE3425A4A9D11311D09FCE24611955F3E@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
Bill et al,
Aspect ratio can be derived from the aspect_ratio_idc syntax element.
When there is a fixed frame rate (i.e., when fixed_frame_rate_flag is
equal to 1, as will be required in some applications), frame rate can be
derived from the num_units_in_tick and time_scale syntax elements.
There are several ways to identify the start of a new picture.  One is
to decode each slice and notice when all macroblocks have been decoded.
When they are present (as will be required in some applications), the
Access unit delimiter RBSP syntax structures will provide easy
identification of the start of each new picture.  Otherwise, the
presence of any SEI message can indicate the start of a new picture and
thus the end of the previous picture.  There are some other syntax
elements that can be used in a similar fashion.  Otherwise, see sections
7.4.1.2.4 ("Detection of the first VCL NAL unit of a primary coded
picture") and 7.4.1.2.5 ("Order of VCL NAL units and association to
coded pictures") of the standard .
Some of those things may also be supported at the systems level in
certain uses.
Regarding the "identifier of a picture", I would suggest reading the
parts of the standard that relate to the frame_num syntax element and
the variables FrameNum, FrameNumWrap, PicNum, LongTermFrameIdx, and
LongTermPicNum.  And perhaps idr_pic_id.  I don't really know what
purpose you're thinking of in asking this question.  There are other
features that can be used by systems as picture identifiers if desired
-- for example, there is the full-frame snapshot SEI message.
Best Regards,
Gary Sullivan
________________________________
	From: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Song
	Sent: Tuesday, November 15, 2023 5:57 PM
	To: mp4-tech@lists.mpegif.org
	Subject: [Mp4-tech]some puzzles about H264 stream.
	Dear Experts,
	I am new about H264, I am wondering that how to find the
information such as Aspect Ratio, Frame Rate from a H.264 elementary
stream? I didn't find the related syntax elements. 
	The second question is how to find the Pictures end? In H.264,
slice structure is the top level syntax about a picture, how to know the
slice is the new starting of next picture?
	Are there any identifier of a picture? or What syntax element
can be used as picture ID? that is to say, the element or some number
can denote a picture uniquely?
	Thank you in advance,
	Regards.
	Bill
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051116/6a9ac0c5/attachment.html
From samuel  dc.fi.udc.es  Thu Nov 17 08:39:25 2005
From: samuel  dc.fi.udc.es (Samuel Rivas)
Date: Thu Nov 17 13:37:07 2005
Subject: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS
In-Reply-To: <20051115095758min-mxa@mail.jp.nec.com>
References: <20051115095758min-mxa@mail.jp.nec.com>
Message-ID: <20051117073925.GA26208@crusher.lfcia.pri>
min-mxa@cnt.necst.nec.co.jp wrote:
> I am trying to multiplex a MPEG4 ASP video ES into a MPEG2 TS.
> As you know B VOPs are allowed in MPEG4 ASP.
> The MPEG4 ASP video ES is generated by a encoder board and 
> when each AU is being outputed the responding PTS and DTS can be geted.
> So when a I or P VOP is packed into PES packet,how shall I set the DTS 
> in the PESHeader?
  As for MPEG2 ES, the timestamps must be set in the first PES packet
carrying data for the AU that timestamp refers to. The main
difference is that MPEG4 stuff is supposed to be packetized in a SL
packetized stream, but usually the SL header is empty to avoid
redundancy with the PES header. This means that there are tighter
restrictions on how the VOPs are mapped into the PES packets, but I
guess this problem is solved by your encoder.
  Best regards.
-- 
	Samuel
From zander  DResearch.de  Thu Nov 17 11:21:55 2005
From: zander  DResearch.de (Zander, Gunther)
Date: Thu Nov 17 13:37:14 2005
Subject: AW: [Mp4-tech] List of MPEG-4 AVC/H.264 video codecs
Message-ID: <05314F0968443041AF5B30B523CDCCFB10EDC1@hiob.intern.dresearch.de>
Hello,
did you consider to add the t264 codec to your list?
It's an open source H.264 implementation.
http://sourceforge.net/projects/t264
Best Regards,
Gunther Zander
> -----Urspr¨¹ngliche Nachricht-----
> Von: mp4-tech-bounces@lists.mpegif.org 
> [mailto:mp4-tech-bounces@lists.mpegif.org] Im Auftrag von 
> Dmitriy Vatolin
> Gesendet: Donnerstag, 10. November 2005 23:16
> An: mp4-tech@lists.mpegif.org
> Betreff: [Mp4-tech] List of MPEG-4 AVC/H.264 video codecs
> 
> Hello!
> 
>    See below updated list of 61 companies and groups working on
>    H.264/AVC codecs.
> 
>    Updates (for 2 month):
>    + KDDI R&D Labs
>    + Hughes Network Systems, LLC
>    + Conexant Systems, Inc.
>    + Sigma Designs
>    + Cradle Technologies, Inc.
>    + STMicroelectronics
>    + PacketVideo Corp.
>    + SONY Video Editor Vegas 6c with H.264/AVC support
>    + PowerEncoder MPEG-4 AVC
>    + Imagicity optimised H.264/AVC codec
>    + Image Converter 2 Plus
>    + some fixes
> 
>    King thanks for your new info messages to me!
>    
>    Site expanded copy of this list available from
>    http://www.compression-links.info/MPEG-4_AVC_H264
> 
> ==============================================================
> ==============
> LIST of MPEG-4 AVC/H.264 VIDEO CODECS
> (Last updated 10 November 2023 - 61 items now!) 
> ==============================================================
> ==============
> 
>     ----------------------------------------------------------
>     SOFTWARE CODECS:
>     ---------------------
> 
>    1. Ateme H.264/MPEG-4 AVC Codec (Ateme)
>       No download on original site:
>       http://www.ateme.com/products/h264.php
>       See 30 day trial of Nero Digital (with Nero Recode) 
> available from:
>       http://www.nerodigital.com/eng/index.html
>       ---
>    2. VSS H.264 Video Codec (Vanguard Software Solutions, Inc.)
>       Trial available, VfW.
>       http://www.vsofts.com/h264/codecs.html
>       ---
>    3. Mainconcept H.264 (MainConcept AG)
>       Trial available on download page
>       http://www.mainconcept.com/h264_encoder.shtml
>       Mainconcept declare partnership with Elecard in H.264 
> in March 2005.
>       ---
>    4. Elecard OneClick Compressor & Elecard AVC/H.264 Decoder 
> Package (Elecard)
>       21-day free trial. Have H.264 SDK
>       http://www.elecard.com/products/oneclick.shtml
>       ---
>    5. Moonlight H.264 Video Codec (Moonlight Cordless LTD.)
>       21-day evaluation, DirectX filter.
>       http://www.moonlight.co.il/products/consumer/oneclickavc/
>       This codec was developed with Elecard team.
>       ---
>    6. SONY Video Editor Vegas 6c with AVC support (Sony Corp.) NEW!!!
>       30-days evaluation.
>       
> http://www.sonymediasoftware.com/products/showproduct.asp?PID=965
>       ---
>    7. QuickTime 7 H.264 (Apple Computer, Inc.)
>       Integrated. Who test it? Any remarks welcome!
>       http://www.apple.com/quicktime/technologies/h264/
>       ---
>    8. FastVDO H.264 (FastVDO LLC)
>       Demo has decode time limit of 5 min.
>       http://www.fastvdo.com/H.264.html
>       ---
>    9. LEAD H.264 Video codec (LEAD Technologies)
>       Only demo, but you can buy codec online (100$). Who test it?
>       http://www.leadcodecs.com/Codecs/LEAD-H264.htm
>       ---
>   10. Compression Master 3 (Popwire/Teleca AB)
>       H264 encoder is all a part of Popwire's Compression Master.
>       Demo soft available (encoding is limited to 20 seconds).
>       http://www.popwire.com/product_info.php?cPath=1&products_id=2
>       ---
>   11. SVM H.264 Decoder Kit & MKi DVD Converter (Pegasus 
> Information Technology Inc.)
>       H.264 encoder and decoder with non standart format.
>       http://www.h-264.com/downloads.htm
>       ---
>   12. AVC Alliance free demo player (AVC Alliance, written by 
> Philips Electronics)
>       Poor but free.
>       http://www.avc-alliance.nl/main/downloads.htm
>       ---
>   13. Intel IPP H.264 codec (Intel Corporation)
>       Intel IPP Library containe now H.264 encoder and decoder.
>       Higher quality codecs are promised in 5.0 version (released now)
>       
> http://www.intel.com/cd/software/products/asmo-na/eng/perflib/
> ipp/index.htm
>       ---
>   14. PowerEncoder MPEG-4 AVC (CyberLink Corp.) NEW!!!
>       Available in online shop.
>       http://www.cyberlink.com/multi/products/main_43_ENU.html
>       ---
>   15. ATI H.264 (ATI Technologies Inc.)
>       No download. Software only and +HW support. Very fast 
> codec by our tests.
>       http://www.ati.com/technology/h264.html
>       ---
>   16. Mpegable AVC Codec (dicas digital image coding GmbH)
>       No download, VfW
>       http://www.mpegable.com/show/mpegableavc.html
>       Old version from free-codecs:
>       
> http://www.codec-download.com/modules.php?name=Downloads&d_op=
> getit&lid=436
>       ---
>   17. Expert H.264 (PixelTools Corporation)
>       Download on request
>       http://www.pixeltools.com/experth264.html
>       ---
>   18. Softstream H.264/MPEG-4 (Media Excel)
>       Audio & Video encoders ane decoders. Evaluation copy on request
>       http://www.mediaexcel.com/products.htm
>       ---
>   19. MPEG-2/HDV/H.264 software (KDDI R&D Labs. Inc.)
>       No download. MPEG related products also with H.264 support
>       http://avs.kddilabs.jp/mpeg/indexe.html
>       ---
>   20. Fraunhofer IIS H.264 Codec (Fraunhofer IIS)
>       No download.
>       http://www.iis.fraunhofer.de/amm/download/mpeg4/
>       ---
>   21. UBLive-264-C64 (UB Video Incorporated)
>       Demo available on request.
>       http://www.ubvideo.com/mainmenu.html
>       ---
>   22. Sorenson Squeeze 4 Compression Suite (Sorenson)
>       No download. Converter.
>       http://www.sorensonmedia.com/solutions/prod/mpg4_win.php
>       ---
>   23. Sonic¨ˆs HD-Series AVC encoder (Sonic Solutions)
>       No download. Only press release now.
>       http://www.sonic.com/about/press/news/2005/04/hdseries.aspx
>       ---
>   24. Enchansed sklmp4 (Pascal Massimino)
>       No download. Announce of 264 development.
>       http://skal.planet-d.net/coding/mpeg4codec.html
>       ---
>   25. NEX VISION H.264 (NEX VISION)
>       No download.
>       http://www.nexvision.fr/
>       ---
>   26. Hughes Network Systems H.264 (Hughes Network Systems, LLC)
>       No download. Own Win&Linux codec for internal usage
>       
> http://www.hns.com/HNS/Doc/0/MGD6UC9UN984BDH4VOK3N53S37/03-03-
> 04_IPoS.html
>       ---
>   27. Imagicity H.264/AVC codec (Imagicity, UK) NEW!!!
>       No download. Speed optimization claimed.
>       http://www.imagicity.com/products/products.htm
>       ---
> 
>       ----------------------------------------------------------
>       OPEN SOURCE:
>       ---------------------
>   28. x264 (free, supported by VideoLAN)
>       OPEN SOURCE. Good publically available high profile encoder.
>       Developers declare "early development stage", but 
> encode results are one of the best.
>       http://www.videolan.org/x264.html
>       http://x264.nl/
>       ---
>   29. JM 9.6 reference H.264 encodec (free, supported by 
> Joint Video Team ISO+ITU-T)
>       OPEN SOURCE. Can be used only for stream testing.
>       http://bs.hhi.de/~suehring/tml/
>       ---
>   30. libavcodec H.264 Decoder (Open Source Project FFMPEG)
>       OPEN SOURCE. Permanently updated decoder.
>       http://ffmpeg.sourceforge.net/index.php
>       ---
>   31. Hdot264 Pre-Alpha (Open Source Project Hdot264)
>       OPEN SOURCE.
>       http://sourceforge.net/projects/hdot264/
>       ---
>   32. Minimal H.264 decoder (by Martin Fiedler)
>       OPEN SOURCE. Basic H.264 decoder.
>       http://www-user.tu-chemnitz.de/~mfie/stuff/SA/
>       ---
> 
>       See also Software Developments Kits for software codec usage:
>       ---
>          1. Elecard AVC&MPEG2 Encoder SDK (Elecard)
>             MPEG2 and H.264 Video Encoder and Decoders other
>             components within the Microsoft¨a DirectShow¨a technology.
>             http://www.elecard.com/products/encoder_sdk.shtml
>             http://www.elecard.com/products/decoder_sdk.shtml
>          2. MainConcept H.264 SDK (MainConcept AG)
>             Windows, Apple, Linux and TI DSP supported.
>             http://developer.mainconcept.com/h264-sdk.html
>          3. LEADTOOLS Multimedia Developer SDKs (LEAD 
> Technologies, Inc)
>             Support for many-many formats, including H.264
>             
> http://www.leadtools.com/SDK/Multimedia/Multimedia-Products-n.htm
>          4. MPEG SDK "MP-Factory" (KDDI R&D Labs. Inc.)
>             SDK with H.264, HDV, MPEG-1,-2, and -4 support
>             http://avs.kddilabs.jp/mpeg/mpfs32/indexe.html
>          5. VSS H.264 SDK (Vanguard Software Solutions, Inc.)
>             Based on there codec. Windows and Linux support.
>             http://www.vsofts.com/h264/sdk.html
>          6. Pegasus H.264 SDK Development Kits (Pegasus)
>             Looks like simple codec, but prefilters declared.
>             http://www.h-264.com/Products.htm
> 
>       ----------------------------------------------------------
>       HARDWARE IMPLEMENTED:
>       ---------------------
>   33. Sony PlayStation Portable (Sony Computer Entertainment Inc.)
>       Product: Support H.264/MPEG-4 AVC Main Profile Level3 decoding.
>       http://www.us.playstation.com/consoles.aspx?id=4
>       See also Image Converter 2 Plus:
>       
> http://www.jp.sonystyle.com/Nws/Software_dl/Pc/Software/Haa/21
> 05710489100.html
>       ---
>   34. VSX 300 Videoconferencing Terminal with H.264 support 
> (Polycom Inc)
>       Product: Biggest HW videoconferencing company, prepare 
> many solutions with 264 support
>       
> http://www.polycom.com/products_services/1,1443,pw-35-6197,00.html
>       ---
>   35. Tandberg EN5930 (Tandberg Television)
>       Product for videoconferencing. Who test it? I have bad review.
>       http://www.tandbergtv.com/productview.asp?n=65
>       ---
>   36. Tandberg Telecom - video communication (Tandberg Telecom)
>       Product for videoconferencing, supports H.264 baseline profile.
>       http://www.tandberg.net/collateral/white_papers/H264.pdf
>       ---
>   37. 4Caster MPEG-4 / H.264 Encoder (Envivio, Inc.)
>       Product: Hardware box encoder
>       http://www.envivio.com/products/4caster.html
>       http://www.envivio.com/products/4caster_se.html 
> (software, no download)
>       ---
>   38. Tut Systems MPEG-4 AVC (Tut Systems, Inc.)
>       Product: Hardware transcoder box.
>       http://www.tutsys.com/pdflibrary/pdf/MPEG4_AVC_2_9_05.pdf
>       ---
>   39. Hardware MGW 5100 (Optibase, Inc.)
>       Product: Hardware box encoder
>       http://www.optibase.com/Content.aspx?id=38
>       ---
>   40. Harmonic DiviCom¨a MV 100 Encoder (Harmonic Inc. )
>       Product: Hardware box encoder
>       
> http://www.harmonicinc.com/view_csd_product_group.cfm?classID=1010
>       ---
>   41. SkyStream Mediaplex & iPlex (SkyStream Networks , inc.)
>       Product: Hardware box encoder.
>       Platforms contain the AVC Encoder Submodule, press release only
>       http://www.skystream.com/press/2005/041805-mpeg4.asp
>       ---
>   42. E42 - H264 DVB over IP Broadcast Encoder (StreamTel Corporation)
>       Product: Hardware box encoder.
>       Broadcast-Grade real time H264 DVB IP streaming encoder
>       
> http://www.streamtel.com/streamtel/products/det/1_Encoders/46_
> E42_-_H264_DVB_over_IP_Broadcast_Encoder.htm
>       ---
>   43. Darim MV401EXP-H264 Encoder and MV410EXP Decoder (Darim 
> Vision Co., Ltd.)
>       Product: Hardware box Encoding/Streaming systems
>       http://www.darimmedia.com/products/mv401exp-h264.php
>       ---
>   44. Mayah I [io] 8000 series (MAYAH Communications)
>       Product: Hardware box. Plan AVC availability in Q3 '05.
>       http://www.mayah.com/products/io-8000a.htm
>       ---
>   45. Thales ARGOS MPEG-4 Encoder (Thales Broadcast & 
> Multimedia, Inc.)
>       Product: Hardware box. Plan AVC from July 05.
>       
> http://www-db.thomcastcom.ext.coltfrance.com/db/thomcast/webdr
> iver?MIval=V_Prod_Detail_b.html&F_ID=4&D_ID=1&P_ID=221
>       ---
>   46. ViBE MPEG-4 AVC (Thomson)
>       Chip: modular hardware series of encoders
>       
> http://www.thomsongrassvalley.com/products/transmission/vibe/e
> ncoder_mpeg4/
>       ---
>   47. Broadcom chips (Broadcom Corp. former Sandvideo)
>       Chip with H.264 HD / decoding only
>       
> http://www.broadcom.com/products/Cable/HDTV-SDTV-Video,-Graphi
> cs-%26-Receiver-Chips/BCM7411
>       ---
>   48. Sentivision H.264/AVC Decoder (Sentivision, Japan)
>       Chip with H.264 HD / decoding only / only for TI DSP's
>       http://www.sentivision.com/products/h264/h264-en.html
>       ---
>   49. SA AVC Encoder (Scientific Atlanta)
>       Product: Transmission system (for satellite TV)
>       http://www.saeurope.com/products/customerhome.htm
>       ---
>   50. H.264 VLE4000 (LSI Logic Corp. former: VideoLocus)
>       Chips with 264 support
>       http://www.lsilogic.com/products/video_broadcasting/vle4000.html
>       ---
>   51. Modulus Video SDTV, HDTV Encoder (Modulus Video, Inc.)
>       Chips. Declarations only now. License H.264 from LSI Logic.
>       http://modulusvideo.com/main.php?Page=12
>       ---
>   52. BSP 15 (System-on-chip) (Equator Technologies, Inc.)
>       Chips. No exact codec info.
>       http://www.equator.com/solutions/digitalmedia.html
>       ---
>   53. BT Exact H.264 Suite (BT Exact)
>       temporarily not available
>       http://www.btexact.com/technologies/ipprofiles?doc=42895
>       ---
>   54. Surveillance Real Time H.264 Encoders (DSP Research, Inc.)
>       Several chips implementation.
>       http://www.dspr.com/www/products/overview_video.htm
>       http://www.dspr.com/www/support/download/video_download.htm
>       ---
>   55. VP3-PMC / Hardware TwinPEG Pro (Vitec Multimedia)
>       Chips. TMS320DM642TMDSP & etc
>       http://www.vitecmm.com/product.php?id=41
>       ---
>   56. WWComs H.264 Codecs (W&W Communications Inc.)
>       Several chips implementation (TI DM64x, C64xx, Freescale DSPs).
>       http://www.wwcoms.com/products/h264_codec.htm
>       http://www.wwcoms.com/products/codec/bc264.htm
>       ---
>   57. CX2418x H.264 Video Decoder IC (Conexant Systems, Inc.)
>       HW Decoder with Main & High Profile (on CX24182)
>       http://www.conexant.com/products/entry.jsp?id=920
>       ---
>   58. Sigma Designs HD decoder (Sigma Designs, Inc.)
>       HW HD Decoder based on SMP8630 family
>       http://www.sigmadesigns.com/products/SMP8630series.htm
>       ---
>   59. Cradle Technologies H.264 (Cradle Technologies, Inc.)
>       HW Main profile Encoder based on CT3600 MDSP family
>       http://www.cradle.com/products/asw_ct3600_h264.shtml
>       ---
>   60. STMicroelectronics H.264 (STMicroelectronics)
>       HW HD High Profile decoder
>       http://www.st.com/stonline/products/literature/bd/11102.htm
>       ---
>   61. PacketVideo H.264 (PacketVideo Corp.)
>       HW Base Profile encoder/decoder on ARM/OMAP/DM642
>       http://www.packetvideo.com/solutions/
>       ---
>       
>    We plan prepare next MSU H.264 codecs comparison test soon will be
>    available from
>    
> http://compression.ru/video/codec_comparison/mpeg-4_avc_h264_en.html
> 
>    If you know any more information about H.264 codecs, please reply
>    this message or add information to Compression-links.info 
> H.264 page:
>    http://www.compression-links.info/MPEG-4_AVC_H264
> 
> ==============================================================
> ==============
>    
> --
> Best regards,
>  Dmitriy Vatolin 
>   Ph.D, Head of Video Group                        
>    mailto:dmitriy@graphics.cs.msu.ru
> 
> 
> 
> 
> _______________________________________________
> NOTE: Please use clear subject lines for your posts. Include 
> [audio, [video], [systems], [general] or another apppropriate 
> identifier to indicate the type of question you have.
> 
> Note: Conduct on the mailing list is subject to the Antitrust 
> guidelines found at 
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Anti
> trust.php
> 
> 
> 
From sma  com.dtu.dk  Thu Nov 17 12:50:08 2005
From: sma  com.dtu.dk (Shankar Manuel Aghito)
Date: Thu Nov 17 13:37:20 2005
Subject: [Mp4-tech] [video] robust H.264 decoder with data partitioning
Message-ID: <B7989DEC7B60254BBC6EF5DE01E312630111F0CB@mail.com.dtu.dk>
I need to simulate UEP with data partitioning on H.264.
Unfortunately, with data partitioning the reference software decoder is
not robust when single partitions are lost.
Could anyone provide me a fixed version of the reference software that
can do that?
Or indicate me another decoder? An older version of the reference
software? A specific commercial product?
Shankar Manuel Aghito
Research Center COM
Technical University of Denmark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051117/48270a24/attachment.html
From sh  upcsurpass.com  Thu Nov 17 20:56:29 2005
From: sh  upcsurpass.com (Song)
Date: Thu Nov 17 13:37:25 2005
Subject: [Mp4-tech]some puzzles about H264 stream.
References: <91D7F2CEE3425A4A9D11311D09FCE24611955F3E@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
Message-ID: <005e01c5eb76$605c57e0$6801a8c0@hongsong>
Dear Gary Sullivan,
Thanks for your kindly help. 
The reason I am looking for the identifier of pictures is that I will calculate the PTS, presentation time stamp, from frame rate and ID of picture.  Because I didn't find the PTS or temporal_reference as MPEG2 have in H.264. What is your suggestion about getting the PTS value of each frame?
Regards.
Bill
  ----- Original Message ----- 
  From: Gary Sullivan 
  To: Song ; mp4-tech@lists.mpegif.org 
  Sent: Thursday, November 17, 2023 2:57 AM
  Subject: RE: [Mp4-tech]some puzzles about H264 stream.
  Bill et al,
  Aspect ratio can be derived from the aspect_ratio_idc syntax element.
  When there is a fixed frame rate (i.e., when fixed_frame_rate_flag is equal to 1, as will be required in some applications), frame rate can be derived from the num_units_in_tick and time_scale syntax elements.
  There are several ways to identify the start of a new picture.  One is to decode each slice and notice when all macroblocks have been decoded.  When they are present (as will be required in some applications), the Access unit delimiter RBSP syntax structures will provide easy identification of the start of each new picture.  Otherwise, the presence of any SEI message can indicate the start of a new picture and thus the end of the previous picture.  There are some other syntax elements that can be used in a similar fashion.  Otherwise, see sections 7.4.1.2.4 ("Detection of the first VCL NAL unit of a primary coded picture") and 7.4.1.2.5 ("Order of VCL NAL units and association to coded pictures") of the standard .
  Some of those things may also be supported at the systems level in certain uses.
  Regarding the "identifier of a picture", I would suggest reading the parts of the standard that relate to the frame_num syntax element and the variables FrameNum, FrameNumWrap, PicNum, LongTermFrameIdx, and LongTermPicNum.  And perhaps idr_pic_id.  I don't really know what purpose you're thinking of in asking this question.  There are other features that can be used by systems as picture identifiers if desired -- for example, there is the full-frame snapshot SEI message.
  Best Regards,
  Gary Sullivan
----------------------------------------------------------------------------
    From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Song
    Sent: Tuesday, November 15, 2023 5:57 PM
    To: mp4-tech@lists.mpegif.org
    Subject: [Mp4-tech]some puzzles about H264 stream.
    Dear Experts,
    I am new about H264, I am wondering that how to find the information such as Aspect Ratio, Frame Rate from a H.264 elementary stream? I didn't find the related syntax elements. 
    The second question is how to find the Pictures end? In H.264, slice structure is the top level syntax about a picture, how to know the slice is the new starting of next picture?
    Are there any identifier of a picture? or What syntax element can be used as picture ID? that is to say, the element or some number can denote a picture uniquely?
    Thank you in advance,
    Regards.
    Bill
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051117/4437ff56/attachment.html
From mykerekes  eml.cc  Thu Nov 17 05:45:17 2005
From: mykerekes  eml.cc (hungary)
Date: Fri Nov 18 13:30:17 2005
Subject: [Mp4-tech] CABAC versus adaptive huffmann coding
In-Reply-To: <1132062730.17268.247550154@webmail.messagingengine.com>
References: <200511141414.jAEEAOLU023641@lists1.magma.ca>
	<71c3c4ac0511142012y7f2e052ei54e90b6088b5126a@mail.gmail.com>
	<1132062730.17268.247550154@webmail.messagingengine.com>
Message-ID: <1132235117.25566.247736813@webmail.messagingengine.com>
Hai experts,
I am again putting this question .....
can any one put some light on 
how CABAC differs from adaptive huffmann coding?
Or both of them are same ?
-- 
  hungary
  mykerekes@eml.cc
-- 
http://www.fastmail.fm - Email service worth paying for. Try it for free
From nikkor  rambler.ru  Thu Nov 17 17:36:10 2005
From: nikkor  rambler.ru (Korotkov Nick)
Date: Fri Nov 18 13:30:23 2005
Subject: [Mp4-tech] MPEG-4 conformance testing???
Message-ID: <web-12147664@mail11.rambler.ru>
Hi all!
Could you help me find files for conformance testing MPEG 4 Video?
Regards,
Nick
From jc  sj.co.uk  Thu Nov 17 16:42:11 2005
From: jc  sj.co.uk (John Cox)
Date: Fri Nov 18 13:30:29 2005
Subject: [Mp4-tech]some puzzles about H264 stream.
In-Reply-To: <005e01c5eb76$605c57e0$6801a8c0@hongsong>
References: <91D7F2CEE3425A4A9D11311D09FCE24611955F3E@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
	<005e01c5eb76$605c57e0$6801a8c0@hongsong>
Message-ID: <0ncpn1hk65egagqv7vb5pud14ad7emh89p@4ax.com>
Bill,
In the general case you can't do it as there is no requirement for the
stream to carry any timing info whatsoever.  (Likewise the aspect ratio
info is optional.)
Assuming that you have the frame rate info in the stream then you still
need to reconcile display order with decode order. Most streams I've met
have Picture Order Counts (POCs) that increase linearly with display
order but this isn't guaranteed either.  The only guaranteed way to
determine output order is to implement enough of a decoder to do the
various picture reordering ops and see what comes out the far end.
On the other hand you might have a stream which contains picture timing
SEI messages with the clock_timestamp_flag set in which case you have
everything you need.
Sorry this isn't very helpful but it seems that the standards committee
decided to divorce the picture coding from ancillary info such as timing
and to make the latter optional.
John Cox
SJ Consulting
>Dear Gary Sullivan,
>
>Thanks for your kindly help. 
>
>The reason I am looking for the identifier of pictures is that I will calculate the PTS, presentation time stamp, from frame rate and ID of picture.  Because I didn't find the PTS or temporal_reference as MPEG2 have in H.264. What is your suggestion about getting the PTS value of each frame?
>
>Regards.
>
>Bill
>
>
>
>  ----- Original Message ----- 
>  From: Gary Sullivan 
>  To: Song ; mp4-tech@lists.mpegif.org 
>  Sent: Thursday, November 17, 2023 2:57 AM
>  Subject: RE: [Mp4-tech]some puzzles about H264 stream.
>
>
>  Bill et al,
>
>  Aspect ratio can be derived from the aspect_ratio_idc syntax element.
>
>  When there is a fixed frame rate (i.e., when fixed_frame_rate_flag is equal to 1, as will be required in some applications), frame rate can be derived from the num_units_in_tick and time_scale syntax elements.
>
>  There are several ways to identify the start of a new picture.  One is to decode each slice and notice when all macroblocks have been decoded.  When they are present (as will be required in some applications), the Access unit delimiter RBSP syntax structures will provide easy identification of the start of each new picture.  Otherwise, the presence of any SEI message can indicate the start of a new picture and thus the end of the previous picture.  There are some other syntax elements that can be used in a similar fashion.  Otherwise, see sections 7.4.1.2.4 ("Detection of the first VCL NAL unit of a primary coded picture") and 7.4.1.2.5 ("Order of VCL NAL units and association to coded pictures") of the standard .
>
>  Some of those things may also be supported at the systems level in certain uses.
>
>  Regarding the "identifier of a picture", I would suggest reading the parts of the standard that relate to the frame_num syntax element and the variables FrameNum, FrameNumWrap, PicNum, LongTermFrameIdx, and LongTermPicNum.  And perhaps idr_pic_id.  I don't really know what purpose you're thinking of in asking this question.  There are other features that can be used by systems as picture identifiers if desired -- for example, there is the full-frame snapshot SEI message.
>
>  Best Regards,
>
>  Gary Sullivan
>
>
>
>----------------------------------------------------------------------------
>    From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Song
>    Sent: Tuesday, November 15, 2023 5:57 PM
>    To: mp4-tech@lists.mpegif.org
>    Subject: [Mp4-tech]some puzzles about H264 stream.
>
>
>    Dear Experts,
>
>    I am new about H264, I am wondering that how to find the information such as Aspect Ratio, Frame Rate from a H.264 elementary stream? I didn't find the related syntax elements. 
>
>    The second question is how to find the Pictures end? In H.264, slice structure is the top level syntax about a picture, how to know the slice is the new starting of next picture?
>
>    Are there any identifier of a picture? or What syntax element can be used as picture ID? that is to say, the element or some number can denote a picture uniquely?
>
>    Thank you in advance,
>
>    Regards.
>
>    Bill
From sh  upcsurpass.com  Fri Nov 18 09:32:34 2005
From: sh  upcsurpass.com (Song)
Date: Fri Nov 18 13:30:35 2005
Subject: [Mp4-tech]some puzzles about H264 stream.
References: <91D7F2CEE3425A4A9D11311D09FCE24611955F3E@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
	<005e01c5eb76$605c57e0$6801a8c0@hongsong>
	<0ncpn1hk65egagqv7vb5pud14ad7emh89p@4ax.com>
Message-ID: <002501c5ebe0$01b3cf10$6701a8c0@hongsong>
Dear John,
Thanks for your help. If a stream doesn't contain any time information or frame rate is not fixed, how does a decoder or a player decode and show that stream without time stamp? About this point, how does the standard committee recommend?
Regards.
Bill
----- Original Message ----- 
From: "John Cox" <jc@sj.co.uk>
To: "Song" <sh@upcsurpass.com>; <mp4-tech@lists.mpegif.org>
Sent: Friday, November 18, 2023 12:42 AM
Subject: Re: [Mp4-tech]some puzzles about H264 stream.
Bill,
In the general case you can't do it as there is no requirement for the
stream to carry any timing info whatsoever.  (Likewise the aspect ratio
info is optional.)
Assuming that you have the frame rate info in the stream then you still
need to reconcile display order with decode order. Most streams I've met
have Picture Order Counts (POCs) that increase linearly with display
order but this isn't guaranteed either.  The only guaranteed way to
determine output order is to implement enough of a decoder to do the
various picture reordering ops and see what comes out the far end.
On the other hand you might have a stream which contains picture timing
SEI messages with the clock_timestamp_flag set in which case you have
everything you need.
Sorry this isn't very helpful but it seems that the standards committee
decided to divorce the picture coding from ancillary info such as timing
and to make the latter optional.
John Cox
SJ Consulting
>Dear Gary Sullivan,
>
>Thanks for your kindly help. 
>
>The reason I am looking for the identifier of pictures is that I will calculate the PTS, presentation time stamp, from frame rate and ID of picture.  Because I didn't find the PTS or temporal_reference as MPEG2 have in H.264. What is your suggestion about getting the PTS value of each frame?
>
>Regards.
>
>Bill
>
>
>
>  ----- Original Message ----- 
>  From: Gary Sullivan 
>  To: Song ; mp4-tech@lists.mpegif.org 
>  Sent: Thursday, November 17, 2023 2:57 AM
>  Subject: RE: [Mp4-tech]some puzzles about H264 stream.
>
>
>  Bill et al,
>
>  Aspect ratio can be derived from the aspect_ratio_idc syntax element.
>
>  When there is a fixed frame rate (i.e., when fixed_frame_rate_flag is equal to 1, as will be required in some applications), frame rate can be derived from the num_units_in_tick and time_scale syntax elements.
>
>  There are several ways to identify the start of a new picture.  One is to decode each slice and notice when all macroblocks have been decoded.  When they are present (as will be required in some applications), the Access unit delimiter RBSP syntax structures will provide easy identification of the start of each new picture.  Otherwise, the presence of any SEI message can indicate the start of a new picture and thus the end of the previous picture.  There are some other syntax elements that can be used in a similar fashion.  Otherwise, see sections 7.4.1.2.4 ("Detection of the first VCL NAL unit of a primary coded picture") and 7.4.1.2.5 ("Order of VCL NAL units and association to coded pictures") of the standard .
>
>  Some of those things may also be supported at the systems level in certain uses.
>
>  Regarding the "identifier of a picture", I would suggest reading the parts of the standard that relate to the frame_num syntax element and the variables FrameNum, FrameNumWrap, PicNum, LongTermFrameIdx, and LongTermPicNum.  And perhaps idr_pic_id.  I don't really know what purpose you're thinking of in asking this question.  There are other features that can be used by systems as picture identifiers if desired -- for example, there is the full-frame snapshot SEI message.
>
>  Best Regards,
>
>  Gary Sullivan
>
>
>
>----------------------------------------------------------------------------
>    From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Song
>    Sent: Tuesday, November 15, 2023 5:57 PM
>    To: mp4-tech@lists.mpegif.org
>    Subject: [Mp4-tech]some puzzles about H264 stream.
>
>
>    Dear Experts,
>
>    I am new about H264, I am wondering that how to find the information such as Aspect Ratio, Frame Rate from a H.264 elementary stream? I didn't find the related syntax elements. 
>
>    The second question is how to find the Pictures end? In H.264, slice structure is the top level syntax about a picture, how to know the slice is the new starting of next picture?
>
>    Are there any identifier of a picture? or What syntax element can be used as picture ID? that is to say, the element or some number can denote a picture uniquely?
>
>    Thank you in advance,
>
>    Regards.
>
>    Bill
From min-mxa  cnt.necst.nec.co.jp  Fri Nov 18 13:33:12 2005
From: min-mxa  cnt.necst.nec.co.jp (min-mxa@cnt.necst.nec.co.jp)
Date: Fri Nov 18 13:30:41 2005
Subject: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS
In-Reply-To: <20051117073925.GA26208@crusher.lfcia.pri>
References: <20051117073925.GA26208@crusher.lfcia.pri>
Message-ID: <20051118133312min-mxa@mail.jp.nec.com>
To Samuel Rivas:
Thank you very much for your reply.
But I am just still not quite clear about the PTS and DTS.
The multiplexer is just packet each MPEG4 AU into each PES packet.
And as you said,the timestamp(PTS/DTS) referring to the AU is setted
into the PESHeader.For I/P VOPs is it true that the PTS has to be
different to the DTS?And for B VOPs is it true that the PTS has to be 
the same as the DTS?
Thanks!
2023/11/17 08:39:25 +0100?Samuel Rivas <samuel@dc.fi.udc.es>??????
?Re: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS????????
>min-mxa@cnt.necst.nec.co.jp wrote:
>> I am trying to multiplex a MPEG4 ASP video ES into a MPEG2 TS.
>> As you know B VOPs are allowed in MPEG4 ASP.
>> The MPEG4 ASP video ES is generated by a encoder board and 
>> when each AU is being outputed the responding PTS and DTS can be geted.
>> So when a I or P VOP is packed into PES packet,how shall I set the DTS 
>> in the PESHeader?
> 
>  As for MPEG2 ES, the timestamps must be set in the first PES packet
>carrying data for the AU that timestamp refers to. The main
>difference is that MPEG4 stuff is supposed to be packetized in a SL
>packetized stream, but usually the SL header is empty to avoid
>redundancy with the PES header. This means that there are tighter
>restrictions on how the VOPs are mapped into the PES packets, but I
>guess this problem is solved by your encoder.
>
>  Best regards.
>-- 
>	Samuel
>_______________________________________________
>NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropri
ate identifier to indicate the type of question you have.
>
>Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/document
s/vault/mp-out-30042-Antitrust.php
From dengxiongshu  yulong.com  Fri Nov 18 14:19:08 2005
From: dengxiongshu  yulong.com (=?utf-8?B?6YKT6ZuE5Lmm?=)
Date: Fri Nov 18 13:30:47 2005
Subject: [Mp4-tech] movie flagment in mp4 file format!
Message-ID: <002401c5ec08$23af1f20$1a040180@ccdomain.com>
Dear mp4 file expert:
    I was developing a mobile recording project. I plan to use movie fragmented file format as ISO14496-12 say that fitted for recording.  But I was confused by some movie fragment box. What mean base-data-offset field in Tfhd box ? And I find it map to initial track's stco box  in Gpac project, so it is mean media data's absolute offset in  movie fragment ? But specification say it  provides an explicit anchor for the data offsets in each track run, I don't know whether it is media data's offset or meta data's offset. And sample-duration field in  trun box  are map to stts box in initial track and  it mean sample delta? Trun box's field sample-size are map to stsz box in initial track? There have been done so in gpac's project, is it right? But I can't find an player that recognize mp4 file contained moof box, and somebody  can give me some demo file with movie fragment that some player can recognize? Can expert clear my confused?
 regards
        thanks in advance?
xiongshu, deng?
e_mail?dengxiongshu@yulong.com
        xiongshudeng@hotmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051118/7af00c71/attachment-0001.html
From samuel  dc.fi.udc.es  Fri Nov 18 10:43:38 2005
From: samuel  dc.fi.udc.es (Samuel Rivas)
Date: Fri Nov 18 13:30:53 2005
Subject: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS
In-Reply-To: <20051118133312min-mxa@mail.jp.nec.com>
References: <20051117073925.GA26208@crusher.lfcia.pri>
	<20051118133312min-mxa@mail.jp.nec.com>
Message-ID: <20051118094338.GB10041@riker.dc.fi.udc.es>
min-mxa@cnt.necst.nec.co.jp dixo:
> To Samuel Rivas:
> 
> Thank you very much for your reply.
> But I am just still not quite clear about the PTS and DTS.
> 
> The multiplexer is just packet each MPEG4 AU into each PES packet.
> And as you said,the timestamp(PTS/DTS) referring to the AU is setted
> into the PESHeader.For I/P VOPs is it true that the PTS has to be
> different to the DTS?And for B VOPs is it true that the PTS has to be 
> the same as the DTS?
  It is. In coding order the I and P frames must be anticipated so that
the B frames can be decoded, i.e. when a B frame is received by the
decoder, the I or P reference frames (one of them with a future PTS)
must had been already decoded. For B VOPs the DTS needn't be sent
because it equals the PTS.
  Regards.
-- 
	Samuel
From jc  sj.co.uk  Fri Nov 18 10:40:33 2005
From: jc  sj.co.uk (John Cox)
Date: Fri Nov 18 13:31:00 2005
Subject: [Mp4-tech]some puzzles about H264 stream.
In-Reply-To: <002501c5ebe0$01b3cf10$6701a8c0@hongsong>
References: <91D7F2CEE3425A4A9D11311D09FCE24611955F3E@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
	<005e01c5eb76$605c57e0$6801a8c0@hongsong>
	<0ncpn1hk65egagqv7vb5pud14ad7emh89p@4ax.com>
	<002501c5ebe0$01b3cf10$6701a8c0@hongsong>
Message-ID: <ilbrn15b2dts68kmbq39it31goqeg9ge9c@4ax.com>
Bill
Personally, if the stream has no timing info then I display it at the
native rate of whatever display device is attached to the output of the
decoder.  But in most cases the video stream is carried in a transport
(such as MPEG2 TS) that contains the PTSs for both Audio and Video and
the only trick then is to make sure that you attach the right PTS to the
right frame.
John Cox
>Dear John,
>
>Thanks for your help. If a stream doesn't contain any time information or frame rate is not fixed, how does a decoder or a player decode and show that stream without time stamp? About this point, how does the standard committee recommend?
>
>Regards.
>
>Bill
>
>
>
>
>----- Original Message ----- 
>From: "John Cox" <jc@sj.co.uk>
>To: "Song" <sh@upcsurpass.com>; <mp4-tech@lists.mpegif.org>
>Sent: Friday, November 18, 2023 12:42 AM
>Subject: Re: [Mp4-tech]some puzzles about H264 stream.
>
>
>Bill,
>
>In the general case you can't do it as there is no requirement for the
>stream to carry any timing info whatsoever.  (Likewise the aspect ratio
>info is optional.)
>
>Assuming that you have the frame rate info in the stream then you still
>need to reconcile display order with decode order. Most streams I've met
>have Picture Order Counts (POCs) that increase linearly with display
>order but this isn't guaranteed either.  The only guaranteed way to
>determine output order is to implement enough of a decoder to do the
>various picture reordering ops and see what comes out the far end.
>
>On the other hand you might have a stream which contains picture timing
>SEI messages with the clock_timestamp_flag set in which case you have
>everything you need.
>
>Sorry this isn't very helpful but it seems that the standards committee
>decided to divorce the picture coding from ancillary info such as timing
>and to make the latter optional.
>
>John Cox
>SJ Consulting
>
>>Dear Gary Sullivan,
>>
>>Thanks for your kindly help. 
>>
>>The reason I am looking for the identifier of pictures is that I will calculate the PTS, presentation time stamp, from frame rate and ID of picture.  Because I didn't find the PTS or temporal_reference as MPEG2 have in H.264. What is your suggestion about getting the PTS value of each frame?
>>
>>Regards.
>>
>>Bill
>>
>>
>>
>>  ----- Original Message ----- 
>>  From: Gary Sullivan 
>>  To: Song ; mp4-tech@lists.mpegif.org 
>>  Sent: Thursday, November 17, 2023 2:57 AM
>>  Subject: RE: [Mp4-tech]some puzzles about H264 stream.
>>
>>
>>  Bill et al,
>>
>>  Aspect ratio can be derived from the aspect_ratio_idc syntax element.
>>
>>  When there is a fixed frame rate (i.e., when fixed_frame_rate_flag is equal to 1, as will be required in some applications), frame rate can be derived from the num_units_in_tick and time_scale syntax elements.
>>
>>  There are several ways to identify the start of a new picture.  One is to decode each slice and notice when all macroblocks have been decoded.  When they are present (as will be required in some applications), the Access unit delimiter RBSP syntax structures will provide easy identification of the start of each new picture.  Otherwise, the presence of any SEI message can indicate the start of a new picture and thus the end of the previous picture.  There are some other syntax elements that can be used in a similar fashion.  Otherwise, see sections 7.4.1.2.4 ("Detection of the first VCL NAL unit of a primary coded picture") and 7.4.1.2.5 ("Order of VCL NAL units and association to coded pictures") of the standard .
>>
>>  Some of those things may also be supported at the systems level in certain uses.
>>
>>  Regarding the "identifier of a picture", I would suggest reading the parts of the standard that relate to the frame_num syntax element and the variables FrameNum, FrameNumWrap, PicNum, LongTermFrameIdx, and LongTermPicNum.  And perhaps idr_pic_id.  I don't really know what purpose you're thinking of in asking this question.  There are other features that can be used by systems as picture identifiers if desired -- for example, there is the full-frame snapshot SEI message.
>>
>>  Best Regards,
>>
>>  Gary Sullivan
>>
>>
>>
>>----------------------------------------------------------------------------
>>    From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Song
>>    Sent: Tuesday, November 15, 2023 5:57 PM
>>    To: mp4-tech@lists.mpegif.org
>>    Subject: [Mp4-tech]some puzzles about H264 stream.
>>
>>
>>    Dear Experts,
>>
>>    I am new about H264, I am wondering that how to find the information such as Aspect Ratio, Frame Rate from a H.264 elementary stream? I didn't find the related syntax elements. 
>>
>>    The second question is how to find the Pictures end? In H.264, slice structure is the top level syntax about a picture, how to know the slice is the new starting of next picture?
>>
>>    Are there any identifier of a picture? or What syntax element can be used as picture ID? that is to say, the element or some number can denote a picture uniquely?
>>
>>    Thank you in advance,
>>
>>    Regards.
>>
>>    Bill
From zhang_lsv  yahoo.com  Fri Nov 18 06:51:27 2005
From: zhang_lsv  yahoo.com (Zhang Y)
Date: Sat Nov 19 18:07:38 2005
Subject: [Mp4-tech] CABAC versus adaptive huffmann coding
In-Reply-To: <1132235117.25566.247736813@webmail.messagingengine.com>
Message-ID: <20051118145132.47376.qmail@web54503.mail.yahoo.com>
HI, hungary,
  It is the difference between arithmetic coding and huffmann coding ((CABAC-- context-adaptive binary arithmetic coding). There are definitely not the same. You can find more details on any data or video compression book. Generally the arithmetic method is more efficient for entropy coding with the complexity tradeoff.
  hope it helps.
  zhang
hungary <mykerekes@eml.cc> wrote:
  Hai experts,
I am again putting this question .....
can any one put some light on 
how CABAC differs from adaptive huffmann coding?
Or both of them are same ?
-- 
hungary
mykerekes@eml.cc
-- 
http://www.fastmail.fm - Email service worth paying for. Try it for free
_______________________________________________
NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have.
Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
---------------------------------
 Yahoo! FareChase - Search multiple travel sites in one click.  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051118/0c6500cc/attachment.html
From singer  apple.com  Fri Nov 18 16:10:05 2005
From: singer  apple.com (Dave Singer)
Date: Sat Nov 19 18:14:43 2005
Subject: [Mp4-tech]some puzzles about H264 stream.
In-Reply-To: <002501c5ebe0$01b3cf10$6701a8c0@hongsong>
References: <91D7F2CEE3425A4A9D11311D09FCE24611955F3E@WIN-MSG-10.wingroup.windeploy.n
	tdev.microsoft.com> <005e01c5eb76$605c57e0$6801a8c0@hongsong>
	<0ncpn1hk65egagqv7vb5pud14ad7emh89p@4ax.com>
	<002501c5ebe0$01b3cf10$6701a8c0@hongsong>
Message-ID: <p06230904bfa39ea14bb9@[81.254.68.167]>
At 9:32  +0800 18/11/05, Song wrote:
>Dear John,
>
>Thanks for your help. If a stream doesn't contain any time 
>information or frame rate is not fixed, how does a decoder or a 
>player decode and show that stream without time stamp? About this 
>point, how does the standard committee recommend?
>
>Regards.
>
>Bill
If the stream is used in an environment where the system layer 
provides timing -- as is the case with both the AVC file format (MP4) 
and RTP -- it is redundant and possibly confusing and contradictory 
to have timing also embedded inside the video layer.  Indeed, there 
is some argument that if the time values (as opposed to order) 
themselves are not used in the process of compression or 
decompression, then they ought to be represented in a coding-system 
independent manner and in a common layer across all media types and 
codecs (which indeed is the approach in MP4 and RTP).
However, since this is not the universal design philosophy, the 
standard allows the timing also to be embedded in streams.  This 
covers the cases of 'raw' 264 streams and so on.
>
>
>
>
>----- Original Message -----
>From: "John Cox" <jc@sj.co.uk>
>To: "Song" <sh@upcsurpass.com>; <mp4-tech@lists.mpegif.org>
>Sent: Friday, November 18, 2023 12:42 AM
>Subject: Re: [Mp4-tech]some puzzles about H264 stream.
>
>
>Bill,
>
>In the general case you can't do it as there is no requirement for the
>stream to carry any timing info whatsoever.  (Likewise the aspect ratio
>info is optional.)
>
>Assuming that you have the frame rate info in the stream then you still
>need to reconcile display order with decode order. Most streams I've met
>have Picture Order Counts (POCs) that increase linearly with display
>order but this isn't guaranteed either.  The only guaranteed way to
>determine output order is to implement enough of a decoder to do the
>various picture reordering ops and see what comes out the far end.
>
>On the other hand you might have a stream which contains picture timing
>SEI messages with the clock_timestamp_flag set in which case you have
>everything you need.
>
>Sorry this isn't very helpful but it seems that the standards committee
>decided to divorce the picture coding from ancillary info such as timing
>and to make the latter optional.
>
>John Cox
>SJ Consulting
>
>>Dear Gary Sullivan,
>>
>>Thanks for your kindly help.
>>
>>The reason I am looking for the identifier of pictures is that I 
>>will calculate the PTS, presentation time stamp, from frame rate 
>>and ID of picture.  Because I didn't find the PTS or 
>>temporal_reference as MPEG2 have in H.264. What is your suggestion 
>>about getting the PTS value of each frame?
>>
>>Regards.
>>
>>Bill
>>
>>
>>
>>   ----- Original Message -----
>>   From: Gary Sullivan
>>   To: Song ; mp4-tech@lists.mpegif.org
>>   Sent: Thursday, November 17, 2023 2:57 AM
>>   Subject: RE: [Mp4-tech]some puzzles about H264 stream.
>>
>>
>>   Bill et al,
>>
>>   Aspect ratio can be derived from the aspect_ratio_idc syntax element.
>>
>>   When there is a fixed frame rate (i.e., when 
>>fixed_frame_rate_flag is equal to 1, as will be required in some 
>>applications), frame rate can be derived from the num_units_in_tick 
>>and time_scale syntax elements.
>>
>>   There are several ways to identify the start of a new picture. 
>>One is to decode each slice and notice when all macroblocks have 
>>been decoded.  When they are present (as will be required in some 
>>applications), the Access unit delimiter RBSP syntax structures 
>>will provide easy identification of the start of each new picture. 
>>Otherwise, the presence of any SEI message can indicate the start 
>>of a new picture and thus the end of the previous picture.  There 
>>are some other syntax elements that can be used in a similar 
>>fashion.  Otherwise, see sections 7.4.1.2.4 ("Detection of the 
>>first VCL NAL unit of a primary coded picture") and 7.4.1.2.5 
>>("Order of VCL NAL units and association to coded pictures") of the 
>>standard .
>>
>>   Some of those things may also be supported at the systems level 
>>in certain uses.
>  >
>>   Regarding the "identifier of a picture", I would suggest reading 
>>the parts of the standard that relate to the frame_num syntax 
>>element and the variables FrameNum, FrameNumWrap, PicNum, 
>>LongTermFrameIdx, and LongTermPicNum.  And perhaps idr_pic_id.  I 
>>don't really know what purpose you're thinking of in asking this 
>>question.  There are other features that can be used by systems as 
>>picture identifiers if desired -- for example, there is the 
>>full-frame snapshot SEI message.
>>
>>   Best Regards,
>>
>>   Gary Sullivan
>>
>>
>>
>>----------------------------------------------------------------------------
>>     From: mp4-tech-bounces@lists.mpegif.org 
>>[mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Song
>>     Sent: Tuesday, November 15, 2023 5:57 PM
>>     To: mp4-tech@lists.mpegif.org
>>     Subject: [Mp4-tech]some puzzles about H264 stream.
>>
>>
>>     Dear Experts,
>>
>>     I am new about H264, I am wondering that how to find the 
>>information such as Aspect Ratio, Frame Rate from a H.264 
>>elementary stream? I didn't find the related syntax elements.
>>
>>     The second question is how to find the Pictures end? In H.264, 
>>slice structure is the top level syntax about a picture, how to 
>>know the slice is the new starting of next picture?
>>
>>     Are there any identifier of a picture? or What syntax element 
>>can be used as picture ID? that is to say, the element or some 
>>number can denote a picture uniquely?
>>
>>     Thank you in advance,
>>
>>     Regards.
>>
>>     Bill
>
>_______________________________________________
>NOTE: Please use clear subject lines for your posts. Include [audio, 
>[video], [systems], [general] or another apppropriate identifier to 
>indicate the type of question you have.
>
>Note: Conduct on the mailing list is subject to the Antitrust 
>guidelines found at 
>http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-- 
David Singer
Apple Computer/QuickTime
From garysull  windows.microsoft.com  Fri Nov 18 13:01:10 2005
From: garysull  windows.microsoft.com (Gary Sullivan)
Date: Sat Nov 19 18:21:08 2005
Subject: [Mp4-tech]some puzzles about H264 stream.
Message-ID: <91D7F2CEE3425A4A9D11311D09FCE246119C5BCA@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
John, Bill, et al,
Timing information is supposed to be available from somewhere - and it
does say that in the spec (Annex C: "In order to check conformance of a
bitstream using the HRD, all sequence parameter sets and picture
parameters sets referred to in the VCL NAL units, and corresponding
buffering period and picture timing SEI messages shall be conveyed to
the HRD, in a timely manner, either in the bitstream (by non-VCL NAL
units), or by other means not specified in this Recommendation |
International Standard.").
If that information isn't in the bitstream itself in any form and isn't
provided at the systems level or specified somehow in an
application-specific context, then all bets are off.   Perhaps the
decoder ought to decode at the maximum speed that is allowed for the
profile & level or at the speed the decoder estimates to be the most
likely intended frame rate, but I think there is no clear guidance about
this.
Best Regards,
Gary Sullivan
+> -----Original Message-----
+> From: mp4-tech-bounces@lists.mpegif.org 
+> [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Song
+> Sent: Thursday, November 17, 2023 5:33 PM
+> To: John Cox; mp4-tech@lists.mpegif.org
+> Subject: Re: [Mp4-tech]some puzzles about H264 stream.
+> 
+> Dear John,
+> 
+> Thanks for your help. If a stream doesn't contain any time 
+> information or frame rate is not fixed, how does a decoder 
+> or a player decode and show that stream without time stamp? 
+> About this point, how does the standard committee recommend?
+> 
+> Regards.
+> 
+> Bill
+> 
+> 
+> 
+> 
+> ----- Original Message ----- 
+> From: "John Cox" <jc@sj.co.uk>
+> To: "Song" <sh@upcsurpass.com>; <mp4-tech@lists.mpegif.org>
+> Sent: Friday, November 18, 2023 12:42 AM
+> Subject: Re: [Mp4-tech]some puzzles about H264 stream.
+> 
+> 
+> Bill,
+> 
+> In the general case you can't do it as there is no 
+> requirement for the
+> stream to carry any timing info whatsoever.  (Likewise the 
+> aspect ratio
+> info is optional.)
+> 
+> Assuming that you have the frame rate info in the stream 
+> then you still
+> need to reconcile display order with decode order. Most 
+> streams I've met
+> have Picture Order Counts (POCs) that increase linearly with display
+> order but this isn't guaranteed either.  The only guaranteed way to
+> determine output order is to implement enough of a decoder to do the
+> various picture reordering ops and see what comes out the far end.
+> 
+> On the other hand you might have a stream which contains 
+> picture timing
+> SEI messages with the clock_timestamp_flag set in which case you have
+> everything you need.
+> 
+> Sorry this isn't very helpful but it seems that the 
+> standards committee
+> decided to divorce the picture coding from ancillary info 
+> such as timing
+> and to make the latter optional.
+> 
+> John Cox
+> SJ Consulting
+> 
+> >Dear Gary Sullivan,
+> >
+> >Thanks for your kindly help. 
+> >
+> >The reason I am looking for the identifier of pictures is 
+> that I will calculate the PTS, presentation time stamp, from 
+> frame rate and ID of picture.  Because I didn't find the PTS 
+> or temporal_reference as MPEG2 have in H.264. What is your 
+> suggestion about getting the PTS value of each frame?
+> >
+> >Regards.
+> >
+> >Bill
+> >
+> >
+> >
+> >  ----- Original Message ----- 
+> >  From: Gary Sullivan 
+> >  To: Song ; mp4-tech@lists.mpegif.org 
+> >  Sent: Thursday, November 17, 2023 2:57 AM
+> >  Subject: RE: [Mp4-tech]some puzzles about H264 stream.
+> >
+> >
+> >  Bill et al,
+> >
+> >  Aspect ratio can be derived from the aspect_ratio_idc 
+> syntax element.
+> >
+> >  When there is a fixed frame rate (i.e., when 
+> fixed_frame_rate_flag is equal to 1, as will be required in 
+> some applications), frame rate can be derived from the 
+> num_units_in_tick and time_scale syntax elements.
+> >
+> >  There are several ways to identify the start of a new 
+> picture.  One is to decode each slice and notice when all 
+> macroblocks have been decoded.  When they are present (as 
+> will be required in some applications), the Access unit 
+> delimiter RBSP syntax structures will provide easy 
+> identification of the start of each new picture.  Otherwise, 
+> the presence of any SEI message can indicate the start of a 
+> new picture and thus the end of the previous picture.  There 
+> are some other syntax elements that can be used in a similar 
+> fashion.  Otherwise, see sections 7.4.1.2.4 ("Detection of 
+> the first VCL NAL unit of a primary coded picture") and 
+> 7.4.1.2.5 ("Order of VCL NAL units and association to coded 
+> pictures") of the standard .
+> >
+> >  Some of those things may also be supported at the systems 
+> level in certain uses.
+> >
+> >  Regarding the "identifier of a picture", I would suggest 
+> reading the parts of the standard that relate to the 
+> frame_num syntax element and the variables FrameNum, 
+> FrameNumWrap, PicNum, LongTermFrameIdx, and LongTermPicNum.  
+> And perhaps idr_pic_id.  I don't really know what purpose 
+> you're thinking of in asking this question.  There are other 
+> features that can be used by systems as picture identifiers 
+> if desired -- for example, there is the full-frame snapshot 
+> SEI message.
+> >
+> >  Best Regards,
+> >
+> >  Gary Sullivan
+> >
+> >
+> >
+> >------------------------------------------------------------
+> ----------------
+> >    From: mp4-tech-bounces@lists.mpegif.org 
+> [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Song
+> >    Sent: Tuesday, November 15, 2023 5:57 PM
+> >    To: mp4-tech@lists.mpegif.org
+> >    Subject: [Mp4-tech]some puzzles about H264 stream.
+> >
+> >
+> >    Dear Experts,
+> >
+> >    I am new about H264, I am wondering that how to find 
+> the information such as Aspect Ratio, Frame Rate from a 
+> H.264 elementary stream? I didn't find the related syntax elements. 
+> >
+> >    The second question is how to find the Pictures end? In 
+> H.264, slice structure is the top level syntax about a 
+> picture, how to know the slice is the new starting of next picture?
+> >
+> >    Are there any identifier of a picture? or What syntax 
+> element can be used as picture ID? that is to say, the 
+> element or some number can denote a picture uniquely?
+> >
+> >    Thank you in advance,
+> >
+> >    Regards.
+> >
+> >    Bill
+> 
+> _______________________________________________
+> NOTE: Please use clear subject lines for your posts. Include 
+> [audio, [video], [systems], [general] or another 
+> apppropriate identifier to indicate the type of question you have.
+> 
+> Note: Conduct on the mailing list is subject to the 
+> Antitrust guidelines found at 
+> http://www.mpegif.org/public/documents/vault/mp-out-30042-Ant
itrust.php
+> 
From garysull  windows.microsoft.com  Fri Nov 18 13:28:49 2005
From: garysull  windows.microsoft.com (Gary Sullivan)
Date: Sat Nov 19 18:27:48 2005
Subject: [Mp4-tech] CABAC versus adaptive huffmann coding
Message-ID: <91D7F2CEE3425A4A9D11311D09FCE246119C5C27@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
CABAC is a type of arithmetic coding.  That is a very different animal
than Huffman coding.  It is typically more efficient (in the compression
sense of efficiency being a measure of the expected number of bits
required to represent the input data) than Huffman coding.
For example, when using Huffman coding, the number of bits used for
encoding each particular source symbol is an integer.  That is not
ordinarily true when using arithmetic coding.  In arithmetic coding, a
sequence of bits is produced in a stream-wise fashion to represent a
(possibly long) series of input symbols and it is often not really
possible to cleanly identify which bits came from which input symbols.
I suggest studying some tutorial material on how arithmetic coding works
(papers by Paul Glor Howard might be a good place to start).  It's not
very easy to explain in an understandable fashion.  It involves a
concept of representing the sequence of input symbols as a fraction that
expresses the probability of occurrence of that sequence of symbols and
then sending enough of the leading bits of that fraction to disambiguate
between the possible values of the symbols.  But that statement probably
makes no sense to someone who doesn't already understand it.
It is worth the effort of studying the subject, as the use of arithmetic
coding concepts is probably going to become a more common feature of
future compression designs.
Best Regards,
Gary Sullivan
+> -----Original Message-----
+> From: mp4-tech-bounces@lists.mpegif.org 
+> [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of hungary
+> Sent: Thursday, November 17, 2023 5:45 AM
+> To: mp4-tech@lists.mpegif.org
+> Subject: [Mp4-tech] CABAC versus adaptive huffmann coding
+> 
+> Hai experts,
+> 
+> I am again putting this question .....
+> 
+> can any one put some light on 
+> how CABAC differs from adaptive huffmann coding?
+> 
+> Or both of them are same ?
+> 
+> -- 
+>   hungary
+>   mykerekes@eml.cc
+> 
+> -- 
+> http://www.fastmail.fm - Email service worth paying for. Try 
+> it for free
+> 
+> _______________________________________________
+> NOTE: Please use clear subject lines for your posts. Include 
+> [audio, [video], [systems], [general] or another 
+> apppropriate identifier to indicate the type of question you have.
+> 
+> Note: Conduct on the mailing list is subject to the 
+> Antitrust guidelines found at 
+> http://www.mpegif.org/public/documents/vault/mp-out-30042-Ant
+> itrust.php
+> 
From singer  apple.com  Sun Nov 20 19:24:31 2005
From: singer  apple.com (Dave Singer)
Date: Mon Nov 21 13:18:09 2005
Subject: [Mp4-tech] movie flagment in mp4 file format!
In-Reply-To: <002401c5ec08$23af1f20$1a040180@ccdomain.com>
References: <002401c5ec08$23af1f20$1a040180@ccdomain.com>
Message-ID: <p0623090fbfa66f532b5c@[81.254.68.167]>
At 14:19 +0800 18/11/05, ???Y ? wrote:
I was developing a mobile recording project. I 
plan to use movie fragmented file format as 
ISO14496-12 say that fitted for recording.  But I 
was confused by some movie fragment box. What 
mean base-data-offset field in Tfhd box ? And I 
find it map to initial track's stco box in Gpac 
project, so it is mean media data's absolute 
offset in  movie fragment ?
-> Just like the chunk offsets in regular movies. 
Absolute offset from file beginning.
But specification say it  provides an explicit 
anchor for the data offsets in each track run, I 
don't know whether it is media data's offset or 
meta data's offset.
-> Media data.
And sample-duration field in trun box are map to 
stts box in initial track and it mean sample 
delta?
-> You could build a regular stts from the TRUN sample times, yes.
  Trun box's field sample-size are map to stsz box in initial track?
-> likewise
  There have been done so in gpac's project, is 
it right? But I can't find an player that 
recognize mp4 file contained moof box
-> I have no idea what you mean by an mp4 file in a moof box.
, and somebody can give me some demo file with 
movie fragment that some player can recognize? 
Can expert clear my confused? regards
thanks in advance?@
xiongshu, deng?@
e_mail?F<mailto:dengxiongshu@yulong.com>dengxiongshu@yulong.com
xiongshudeng@hotmail.com
_______________________________________________ 
NOTE: Please use clear subject lines for your 
posts. Include [audio, [video], [systems], 
[general] or another apppropriate identifier to 
indicate the type of question you have.
Note: Conduct on the mailing list is subject to 
the Antitrust guidelines found at 
http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php 
-- 
David Singer
Apple Computer/QuickTime
From min-mxa  cnt.necst.nec.co.jp  Mon Nov 21 15:14:15 2005
From: min-mxa  cnt.necst.nec.co.jp (min-mxa@cnt.necst.nec.co.jp)
Date: Mon Nov 21 13:25:36 2005
Subject: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS
In-Reply-To: <20051118094338.GB10041@riker.dc.fi.udc.es>
References: <20051118094338.GB10041@riker.dc.fi.udc.es>
Message-ID: <20051121151415min-mxa@mail.jp.nec.com>
I see.
thanks a lot!
2023/11/18 10:43:38 +0100?Samuel Rivas <samuel@dc.fi.udc.es>??????
?Re: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS????????
>min-mxa@cnt.necst.nec.co.jp dixo:
>> To Samuel Rivas:
>> 
>> Thank you very much for your reply.
>> But I am just still not quite clear about the PTS and DTS.
>> 
>> The multiplexer is just packet each MPEG4 AU into each PES packet.
>> And as you said,the timestamp(PTS/DTS) referring to the AU is setted
>> into the PESHeader.For I/P VOPs is it true that the PTS has to be
>> different to the DTS?And for B VOPs is it true that the PTS has to be 
>> the same as the DTS?
>
>  It is. In coding order the I and P frames must be anticipated so that
>the B frames can be decoded, i.e. when a B frame is received by the
>decoder, the I or P reference frames (one of them with a future PTS)
>must had been already decoded. For B VOPs the DTS needn't be sent
>because it equals the PTS.
>
>  Regards.
>-- 
>	Samuel
>_______________________________________________
>NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropri
ate identifier to indicate the type of question you have.
>
>Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/document
s/vault/mp-out-30042-Antitrust.php
From Stephane.Pechard  univ-nantes.fr  Mon Nov 21 10:48:51 2005
From: Stephane.Pechard  univ-nantes.fr (=?iso-8859-1?Q?St=E9phane_PECHARD?=)
Date: Mon Nov 21 13:30:21 2005
Subject: [Mp4-tech] [H.264] I-frames in reference codec
Message-ID: <55485.193.52.89.12.1132566531.squirrel@webmail.univ-nantes.fr>
Hi all,
I use the H.264 reference codec (JM 10.1) to encode HDTV sources in main
profile. I would like to use a certain frame structure but I can't succeed
in. I want to have the following structure :
I-P-B-P-B-P-B-P-B-P-B-P-B-I
i.e. 1 B frame inserted and a 13 period of I-frame. But when I put
IntraPeriod=13 and NumberBFrames=1 (with FrameSkip=1), I never get
I-frames, just one IDR following by B-P structures.
Have I a wrong config or misunderstood something ? I couldn't find help
anywhere else.
Thanx in advance
--
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
From student.mpg  gmail.com  Mon Nov 21 16:29:27 2005
From: student.mpg  gmail.com (mk ali)
Date: Mon Nov 21 13:30:27 2005
Subject: [Mp4-tech] adding ac3 audio in mpeg2 program stream!
Message-ID: <ff9b71a20511210259k49caec18qa365b7ef7ca3ed3f@mail.gmail.com>
Gents,
I tried adding ac3 audio( instead of mpeg1-2) in mpeg2 program
stream.Eventhough i can see ac3 packets in the stream , mpeg anlyser is not
finding any audio stream . It just shows video stream.I think am missing
something in the header level. Can someone throw some light on that?
Thanks in advance
sm
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051121/e65f6f86/attachment.html
From Andreas.Schneider  codingtechnologies.com  Mon Nov 21 14:09:15 2005
From: Andreas.Schneider  codingtechnologies.com (Andreas Schneider)
Date: Tue Nov 22 13:13:44 2005
Subject: [Mp4-tech] Regarding conformance testing of HE-AAC
In-Reply-To: <20051116111804.32795.qmail@web60717.mail.yahoo.com>
Message-ID: <OF1A8CDE6D.580ACD13-ONC12570C0.0047EC07-C12570C0.00484225@codingtechnologies.com>
The 5.1 HE-AAC conformance stream can be found here:
ftp://mpaudconf:adif2mp4@ftp.iis.fhg.de/mpeg4audio-conformance/compressedMp4/al_sbr_cm_48_5.1.mp4
Regards,
Andreas
"Deshpande,Vishvanath" <vishu_dash@yahoo.com> wrote on 16.11.2023 
12:18:04:
> Hi,
> Can any one pls tell me where do we get 5.1 ch
> conformance streams for AAC decoder(AAC-Lc and HE
> AAC)?
> 
> Thanks in advance,
> 
> regards,
> -Vishvanath
> 
> 
> 
> ___________________________________________________________ 
> How much free photo storage do you get? Store your holiday 
> snaps for FREE with Yahoo! Photos http://uk.photos.yahoo.com
--
Andreas Schneider, Research Engineer
Coding Technologies GmbH
Deutschherrnstr. 15-19
90429 Nuernberg, Germany
phone: +49 (0) 911 92891 -26
fax: +49 (0) 911 92891 -99
mailto:snd@CodingTechnologies.com
From alexismt  adelphia.net  Mon Nov 21 08:10:06 2005
From: alexismt  adelphia.net (Alexis Michael Tourapis)
Date: Tue Nov 22 13:24:02 2005
Subject: [Mp4-tech] [H.264] I-frames in reference codec
In-Reply-To: <55485.193.52.89.12.1132566531.squirrel@webmail.univ-nantes.fr>
Message-ID: <003701c5eeb6$0a223640$0200a8c0@AlexLap>
Try IntraPeriod=6. The number only accounts for "primary" structure frames
(i.e. excluding intermediate slices). See also the reference software
manual.
Best regards,
Alexis
-----Original Message-----
From: mp4-tech-bounces@lists.mpegif.org
[mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of St?phane PECHARD
Sent: Monday, November 21, 2023 1:49 AM
To: mp4-tech@lists.mpegif.org
Subject: [Mp4-tech] [H.264] I-frames in reference codec
Hi all,
I use the H.264 reference codec (JM 10.1) to encode HDTV sources in main
profile. I would like to use a certain frame structure but I can't succeed
in. I want to have the following structure :
I-P-B-P-B-P-B-P-B-P-B-P-B-I
i.e. 1 B frame inserted and a 13 period of I-frame. But when I put
IntraPeriod=13 and NumberBFrames=1 (with FrameSkip=1), I never get
I-frames, just one IDR following by B-P structures.
Have I a wrong config or misunderstood something ? I couldn't find help
anywhere else.
Thanx in advance
--
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
_______________________________________________
NOTE: Please use clear subject lines for your posts. Include [audio,
[video], [systems], [general] or another apppropriate identifier to indicate
the type of question you have.
Note: Conduct on the mailing list is subject to the Antitrust guidelines
found at
http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
From Stephane.Pechard  univ-nantes.fr  Mon Nov 21 17:38:21 2005
From: Stephane.Pechard  univ-nantes.fr (=?iso-8859-1?Q?St=E9phane_PECHARD?=)
Date: Tue Nov 22 13:30:10 2005
Subject: [Mp4-tech] [H.264] I-frames in reference codec
In-Reply-To: <003701c5eeb6$0a223640$0200a8c0@AlexLap>
References: <55485.193.52.89.12.1132566531.squirrel@webmail.univ-nantes.fr>
	<003701c5eeb6$0a223640$0200a8c0@AlexLap>
Message-ID: <43249.193.52.89.12.1132591101.squirrel@webmail.univ-nantes.fr>
> Try IntraPeriod=6. The number only accounts for "primary" structure frames
> (i.e. excluding intermediate slices).
Yeah, I did that, it's ok for the period. I also had to put IntraBottom=1
because I work with interlaced videos. Otherwise, I don't have the
structure wanted. Is it the best way to do ?
> See also the reference software manual.
I have the version from October, 2004. Is there a more recent version ?
Best regards,
--
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
From jianpengdong  yahoo.com  Mon Nov 21 12:50:20 2005
From: jianpengdong  yahoo.com (James (Jianpeng) Dong)
Date: Tue Nov 22 13:37:25 2005
Subject: [Mp4-tech] [H.264] I-frames in reference codec
In-Reply-To: <55485.193.52.89.12.1132566531.squirrel@webmail.univ-nantes.fr>
Message-ID: <20051121205020.15812.qmail@web30009.mail.mud.yahoo.com>
I am sure that you made a tiny mistake in your configuration file.
  For the given pattern, the period should be 6, rather than 13. In addition, you should make sure that you have enough number of frames to see the next I frame (note that the total number of frames to be coded does not include the B frames).
  Good luck,
  James
St?phane PECHARD <Stephane.Pechard@univ-nantes.fr> wrote:
  Hi all,
I use the H.264 reference codec (JM 10.1) to encode HDTV sources in main
profile. I would like to use a certain frame structure but I can't succeed
in. I want to have the following structure :
I-P-B-P-B-P-B-P-B-P-B-P-B-I
i.e. 1 B frame inserted and a 13 period of I-frame. But when I put
IntraPeriod=13 and NumberBFrames=1 (with FrameSkip=1), I never get
I-frames, just one IDR following by B-P structures.
Have I a wrong config or misunderstood something ? I couldn't find help
anywhere else.
Thanx in advance
--
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
_______________________________________________
NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have.
Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
---------------------------------------------------------
 James (Jianpeng) Dong 
 Ph.D. Student
 Computer Engineering Department 
 Santa Clara University, CA 95000
 USA 
 Phone: 408-718-8965(C), 408-615-7918(H) 
 Web: http://students.engr.scu.edu/~jdong  
---------------------------------------------------------
---------------------------------
 Yahoo! FareChase - Search multiple travel sites in one click.  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051121/4f9d47d5/attachment.html
From dengxiongshu  yulong.com  Tue Nov 22 10:44:47 2005
From: dengxiongshu  yulong.com (=?utf-8?B?6YKT6ZuE5Lmm?=)
Date: Tue Nov 22 13:37:30 2005
Subject: [Mp4-tech] Re:movie flagment in mp4 file format!
Message-ID: <000e01c5ef0e$b3f22a60$1a040180@ccdomain.com>
Dear expert:
Thanks Dave Singer?s reply. I was also confused by data-offset field in trun box, is it also like stco box in regular movies? Can you tell me relationship between base-data-offset in tfhd box and data-offset in trun box? How can stsc box in regular expand from movie frame boxes? It use base-data-offset to make of his chunk offset, sample count in trun box to make of samples-per-chunk in regular stsc box, entry-count are add one, first-chunk are the last chunk in moov added one , and one movie fragment make up of one chunk, that is specification design philosophy,my understand is right ? 
Thanks very much
Xiongshu,deng
Dengxiongshu@yulong.com
dengxiongshu@hotmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051122/70eda2b0/attachment.html
From min-mxa  cnt.necst.nec.co.jp  Tue Nov 22 16:23:28 2005
From: min-mxa  cnt.necst.nec.co.jp (min-mxa@cnt.necst.nec.co.jp)
Date: Tue Nov 22 13:37:36 2005
Subject: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS
In-Reply-To: <5.1.0.14.2.20051121223631.09dc59e8@mail>
References: <5.1.0.14.2.20051121223631.09dc59e8@mail>
Message-ID: <20051122162328min-mxa@mail.jp.nec.com>
2023/11/21 22:49:35 -0500?Gary Hughes <gary@stargate-video.com>??????
?Re: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS????????
>At 01:33 PM 11/18/2005 +0900, min-mxa@cnt.necst.nec.co.jp wrote:
>>To Samuel Rivas:
>>
>>Thank you very much for your reply.
>>But I am just still not quite clear about the PTS and DTS.
>>
>>The multiplexer is just packet each MPEG4 AU into each PES packet.
>>And as you said,the timestamp(PTS/DTS) referring to the AU is setted
>>into the PESHeader.For I/P VOPs is it true that the PTS has to be
>>different to the DTS?And for B VOPs is it true that the PTS has to be 
>>the same as the DTS?
>
>A DTS can only appear in a PES header if a PTS is present and the decode
>time differs from the presentation time. A B VOP would have only a PTS
>(and the decode time is by definition the same as the presentation time).
that is to say, a DTS shall appear in a PES header that refers to an I or P 
VOP with the start code carried in this PES for a ES including B VOPs.
Is is right?
>
>>> 
>>>  As for MPEG2 ES, the timestamps must be set in the first PES packet
>>>carrying data for the AU that timestamp refers to. The main
>>>difference is that MPEG4 stuff is supposed to be packetized in a SL
>>>packetized stream, but usually the SL header is empty to avoid
>>>redundancy with the PES header. This means that there are tighter
>>>restrictions on how the VOPs are mapped into the PES packets, but I
>>>guess this problem is solved by your encoder.
>
>13818-1 supports the carriage of 14496-2 (video) and 14496-3 (audio)
>elementary streams in PES packets, in addition to 14496-1 scenes with
>their associated streams. When carrying elementary streams, only tools
>defined in 13818-1 are used. When carrying 14496-1 scenes, they may be
>SL packetized streams or FlexMux streams. The 2000 edition of 13818-1
>contains all of the amendments related to MPEG-4 in MPEG-2 systems.
>
>gary
>
From mp3.aac.mp4  gmail.com  Tue Nov 22 14:20:29 2005
From: mp3.aac.mp4  gmail.com (tech list)
Date: Tue Nov 22 13:37:43 2005
Subject: [Mp4-tech] Some doubts in video
In-Reply-To: <91D7F2CEE3425A4A9D11311D09FCE246118627CC@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
References: <91D7F2CEE3425A4A9D11311D09FCE246118627CC@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com>
Message-ID: <409a09b90511220050v3a547ac4s42ceff201c7778f1@mail.gmail.com>
Mayank, looks like you went for an interview and rubbed someone the wrong
way!!
Here's my 2 cents worth:
+> > 1.In MPEG-4 simple profile how many motion vectors are
+> > transmitted from encoder to decoder or generalizing
+> > the above can anybody tell how to determine the number
+> > of motion vectors being transmitted from encoder to
+> > decoder.
Not sure really...number of B/P-frames * number of MV per macro block????
+> > 2.Suppose in the video encoding we don't do
+> > quantization then what will be affect on final encoded
+> > stream.
So in this case we can assume that transformation is done, but no
quantization. Ignoring whether this is a valid bitstream or not, the effect
would be that the compression would be lesser. Note that some compression
would be done by virtue of the transform coding itself, coupled with VLC in
the later stages.
+> > 3.Suppose we are encoding QCIF sized video then what
+> > will be the minimum size of decoded buffer.I suppose
+> > it is 176*144+2*176/2*144/2.Please correct me if i am
+> > wrong.
Depends on output colourspace. Your answer is correct for YUV4:2:0, the most
common colourspace expected from a decoder.
Best answer would be VIDEO_WIDTH*VIDEO_HIEGHT*sizeof(COLOUR_SPACE)
+> > 4.Suppose we are doing some video integration and we
+> > found out that DSP memory is not enough for the
+> > encoded stream then how can we increase DSP memory.
Is it encoded stream, or decoded stream??
Simple answer would be that you throw out the damn DSP and get a better one
;-)
If it is encoded stream, then assuming that the raw video is coming from a
camera source, you can start writing into a file as soon as a certain
watermark level is reached.
Seriously, it would be totally dependent on the architecture of the DSP. In
case of TI's OMAP, you can switch from external to internal memories. Here
it looks like this guy had such a problem and he did something in his target
DSP, and he wanted you to guess what that was... tough luck!!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051122/cdce936d/attachment-0001.html
From gripened  gmail.com  Tue Nov 22 14:46:56 2005
From: gripened  gmail.com (Jayant Chauhan)
Date: Tue Nov 22 13:37:50 2005
Subject: [Mp4-tech] [MP4 parser DirectShow Filter]
Message-ID: <5b996acb0511220116u7705982du4c932cf4cc5ee37e@mail.gmail.com>
Hey
Does anyone have source code for MP4 parser's filter other than the one by
3iVX ?!
 regards
Jayant
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051122/fcabe4ea/attachment.html
From mp4-tech-owner  lists.mpegif.org  Tue Nov 22 13:50:31 2005
From: mp4-tech-owner  lists.mpegif.org (MPEGIF List Bounces)
Date: Wed Nov 23 09:21:55 2005
Subject: FW: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS
Message-ID: <003c01c5ef63$52d4b680$0200a8c0@corp.intertrust.com>
Forwarding email from non-subscribed sender address.
-----Original Message-----
From: Gary Hughes [mailto:gary@stargate-video.com] 
Sent: Tuesday, 22 November 2023 04:50
To: min-mxa@cnt.necst.nec.co.jp; Samuel Rivas
Cc: mp4-tech@lists.mpegif.org
Subject: Re: [Mp4-tech] multiplex MPEG4 ASP video ES into MPEG2 TS
At 01:33 PM 11/18/2005 +0900, min-mxa@cnt.necst.nec.co.jp wrote:
>To Samuel Rivas:
>
>Thank you very much for your reply.
>But I am just still not quite clear about the PTS and DTS.
>
>The multiplexer is just packet each MPEG4 AU into each PES packet.
>And as you said,the timestamp(PTS/DTS) referring to the AU is setted
>into the PESHeader.For I/P VOPs is it true that the PTS has to be
>different to the DTS?And for B VOPs is it true that the PTS has to be 
>the same as the DTS?
A DTS can only appear in a PES header if a PTS is present and the decode
time differs from the presentation time. A B VOP would have only a PTS
(and the decode time is by definition the same as the presentation time).
>> 
>>  As for MPEG2 ES, the timestamps must be set in the first PES packet
>>carrying data for the AU that timestamp refers to. The main
>>difference is that MPEG4 stuff is supposed to be packetized in a SL
>>packetized stream, but usually the SL header is empty to avoid
>>redundancy with the PES header. This means that there are tighter
>>restrictions on how the VOPs are mapped into the PES packets, but I
>>guess this problem is solved by your encoder.
13818-1 supports the carriage of 14496-2 (video) and 14496-3 (audio)
elementary streams in PES packets, in addition to 14496-1 scenes with
their associated streams. When carrying elementary streams, only tools
defined in 13818-1 are used. When carrying 14496-1 scenes, they may be
SL packetized streams or FlexMux streams. The 2000 edition of 13818-1
contains all of the amendments related to MPEG-4 in MPEG-2 systems.
gary
From mp4-tech-owner  lists.mpegif.org  Tue Nov 22 17:39:34 2005
From: mp4-tech-owner  lists.mpegif.org (MPEGIF List Bounces)
Date: Wed Nov 23 09:29:17 2005
Subject: FW: [Mp4-tech] [H.264] I-frames in reference codec
Message-ID: <004101c5ef83$53a31c80$0200a8c0@corp.intertrust.com>
forwarding for non-subscribed poster
-----Original Message-----
From: Tourapis, Alexis [mailto:atour@dolby.com] 
Sent: Tuesday, 22 November 2023 15:08
To: St?phane PECHARD; mp4-tech@lists.mpegif.org
Subject: RE: [Mp4-tech] [H.264] I-frames in reference codec
The latest version is JVT-Q042
(http://ftp3.itu.ch/av-arch/jvt-site/2005_10_Nice/JVT-Q042-L.zip). 
What do you mean by best way? 
Best regards,
Alexis
	-----Original Message----- 
	From: mp4-tech-bounces@lists.mpegif.org on behalf of St?phane
PECHARD 
	Sent: Mon 11/21/2005 8:38 AM 
	To: mp4-tech@lists.mpegif.org 
	Cc: 
	Subject: RE: [Mp4-tech] [H.264] I-frames in reference codec
	> Try IntraPeriod=6. The number only accounts for "primary"
structure frames
	> (i.e. excluding intermediate slices).
	Yeah, I did that, it's ok for the period. I also had to put
IntraBottom=1
	because I work with interlaced videos. Otherwise, I don't have the
	structure wanted. Is it the best way to do ?
	> See also the reference software manual.
	I have the version from October, 2004. Is there a more recent
version ?
	Best regards,
	--
	~ st?phane p?chard
	~ phd student | irccyn-ivc | france
	~ http://www.irccyn.ec-nantes.fr/~pechard
	~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
	_______________________________________________
	NOTE: Please use clear subject lines for your posts. Include [audio,
[video], [systems], [general] or another apppropriate identifier to indicate
the type of question you have.
	Note: Conduct on the mailing list is subject to the Antitrust
guidelines found at
http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-----------------------------------------
This message (including any attachments) may contain confidential
information intended for a specific individual and purpose.  If you are
not the intended recipient, delete this message.  If you are not the
intended recipient, disclosing, copying, distributing, or taking any
action based on this message is strictly prohibited.
From stephane.pechard  univ-nantes.fr  Tue Nov 22 19:37:41 2005
From: stephane.pechard  univ-nantes.fr (=?utf-8?q?St=C3=A9phane_P=C3=A9chard?=)
Date: Wed Nov 23 09:36:35 2005
Subject: [Mp4-tech] [H.264] I-frames in reference codec
In-Reply-To: <7272EE229DA1AA48B47EBDC47EB0C68B019D9193@sapphire.dolby.net>
References: <7272EE229DA1AA48B47EBDC47EB0C68B019D9193@sapphire.dolby.net>
Message-ID: <200511221937.52234.stephane.pechard@univ-nantes.fr>
> The latest version is JVT-Q042
> (http://ftp3.itu.ch/av-arch/jvt-site/2005_10_Nice/JVT-Q042-L.zip).
Thanx a lot for that.
> What do you mean by best way?
Well, I was not sure about the use of this option. Actually, I was surprised 
that without it I had no I-frames. So I asked if using this option was the 
right way (for the coder and the resulting file) to obtain the structure I 
want or if it's useless and there's another option I didn't see.
Best regards,
St?phane
>
> 	-----Original Message-----
> 	From: mp4-tech-bounces@lists.mpegif.org on behalf of St?phane PECHARD
> 	Sent: Mon 11/21/2005 8:38 AM
> 	To: mp4-tech@lists.mpegif.org
> 	Cc:
> 	Subject: RE: [Mp4-tech] [H.264] I-frames in reference codec
>
> 	> Try IntraPeriod=6. The number only accounts for "primary" structure
> 	> frames (i.e. excluding intermediate slices).
>
> 	Yeah, I did that, it's ok for the period. I also had to put IntraBottom=1
> 	because I work with interlaced videos. Otherwise, I don't have the
> 	structure wanted. Is it the best way to do ?
>
> 	> See also the reference software manual.
>
> 	I have the version from October, 2004. Is there a more recent version ?
>
> 	Best regards,
>
> 	--
> 	~ st?phane p?chard
> 	~ phd student | irccyn-ivc | france
> 	~ http://www.irccyn.ec-nantes.fr/~pechard
> 	~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
-- 
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 307 bytes
Desc: not available
Url : /pipermail/mp4-tech/attachments/20051122/c08da390/attachment.bin
From Danijel.Domazet  zg.t-com.hr  Tue Nov 22 20:14:41 2005
From: Danijel.Domazet  zg.t-com.hr (Danijel Domazet)
Date: Wed Nov 23 09:36:42 2005
Subject: [Mp4-tech][Audio][HE AAC] 1, 2, or 4 envelopes for FIXFIX frame?
Message-ID: <000d01c5ef98$fdc327e0$5b00a8c0@FERGUSON>
Hi all,
Regarding HE AAC audio... 
There can be 1, 2, or 4 envelopes for the FIXFIX frame (stationary frame -
no transients). How does one decide how many envelopes to use for such a
frame? Sound gets better as there are more envelopes but it takes bitrate
also so one can't use 4 envelopes all the time (just to make this clear).
The concept of variable frame borders in case of transients is pretty
clear - but the case of stationary frame where borders are predifined to be
0-4-8-12-16 slots I don't quite understand. What should I base my decision
on? 1, 2 or 4 envelopes?
Daniel
From kaps_agl  yahoo.co.in  Wed Nov 23 07:19:24 2005
From: kaps_agl  yahoo.co.in (kapil agrawal)
Date: Wed Nov 23 09:36:47 2005
Subject: [Mp4-tech] YUVs at different resolutions
Message-ID: <20051123071924.58322.qmail@web8502.mail.in.yahoo.com>
Hello, 
I am prensently wrkng on H264 and want to do an
statistical analysis  of diffrent YUVs.
So can someone provide me the source for different
YUVs at different resolutions.
Thanks
Kapil 
__________________________________________________________ 
Enjoy this Diwali with Y! India Click here http://in.promos.yahoo.com/fabmall/index.html
From mp4-tech-owner  lists.mpegif.org  Wed Nov 23 10:07:06 2005
From: mp4-tech-owner  lists.mpegif.org (MPEGIF List Bounces)
Date: Wed Nov 23 14:44:19 2005
Subject: [Mp4-tech] [H.264] I-frames in reference codec
In-Reply-To: <7272EE229DA1AA48B47EBDC47EB0C68B01A4DAB0@sapphire.dolby.net>
Message-ID: <006901c5f00d$473f0280$0200a8c0@corp.intertrust.com>
Alexis,
Your sending address is not subscribed ot the list, so the email was
rejected.
Forwarding this once.
Best,
Rob
> -----Original Message-----
> From: Tourapis, Alexis [mailto:atour@dolby.com] 
> Sent: Tuesday, 22 November 2023 21:17
> To: St?phane P?chard; mp4-tech@lists.mpegif.org
> Subject: RE: [Mp4-tech] [H.264] I-frames in reference codec
> 
> The use or not of Intra slices, or even frequent usage of 
> IDRs (note that they are not the same thing although an IDR 
> will be coded using Intra slices) depends on your target 
> application. In some applications where the goal might be the 
> best compression performance possible without caring about 
> random access, has lower requirements in terms of error 
> resilience etc, one could for example insert intra slices 
> where necessary (i.e. during scene chane changes or where 
> performance justifies the use of Intra slices). For your 
> reference the software in fact supports a special RD mode 
> where such a test is performed (i.e. under certain conditions 
> test whether the use of an Intra slice would lead to better 
> performance). Unfortunately at this point this only works for 
> Frame coding but will see if we find time to update it for 
> Interlace coding as well.
> 
> Btw, note that even if you do not use the flag IntraBottom, 
> the encoder will still insert an intra picture (i.e. top 
> field would be intra). However the encoder reports only the 
> second field. In fact, note that some of the statistics when 
> using RD level picture decisions not reported correctly (i.e. 
> last encoding pass data are shown instead of best ones). We 
> should fix this as well.
> 
> Best regards,
> 
> Alexis
> 
> -----Original Message-----
> From: St?phane P?chard [mailto:stephane.pechard@univ-nantes.fr] 
> Sent: Tuesday, November 22, 2023 10:38 AM
> To: Tourapis, Alexis; mp4-tech@lists.mpegif.org
> Subject: Re: [Mp4-tech] [H.264] I-frames in reference codec
> 
> > The latest version is JVT-Q042
> > (http://ftp3.itu.ch/av-arch/jvt-site/2005_10_Nice/JVT-Q042-L.zip).
> 
> Thanx a lot for that.
> 
> > What do you mean by best way?
> 
> Well, I was not sure about the use of this option. Actually, 
> I was surprised that without it I had no I-frames. So I asked 
> if using this option was the right way (for the coder and the 
> resulting file) to obtain the structure I want or if it's 
> useless and there's another option I didn't see.
> 
> Best regards,
> 
> St?phane
> 
> >
> > 	-----Original Message-----
> > 	From: mp4-tech-bounces@lists.mpegif.org on behalf of 
> St?phane PECHARD
> > 	Sent: Mon 11/21/2005 8:38 AM
> > 	To: mp4-tech@lists.mpegif.org
> > 	Cc:
> > 	Subject: RE: [Mp4-tech] [H.264] I-frames in reference codec
> >
> > 	> Try IntraPeriod=6. The number only accounts for 
> "primary" structure
> > 	> frames (i.e. excluding intermediate slices).
> >
> > 	Yeah, I did that, it's ok for the period. I also had to 
> put IntraBottom=1
> > 	because I work with interlaced videos. Otherwise, I 
> don't have the
> > 	structure wanted. Is it the best way to do ?
> >
> > 	> See also the reference software manual.
> >
> > 	I have the version from October, 2004. Is there a more 
> recent version ?
> >
> > 	Best regards,
> >
> > 	--
> > 	~ st?phane p?chard
> > 	~ phd student | irccyn-ivc | france
> > 	~ http://www.irccyn.ec-nantes.fr/~pechard
> > 	~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
> 
> --
> ~ st?phane p?chard
> ~ phd student | irccyn-ivc | france
> ~ http://www.irccyn.ec-nantes.fr/~pechard
> ~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
> 
> 
> -----------------------------------------
> This message (including any attachments) may contain confidential
> information intended for a specific individual and purpose.  
> If you are
> not the intended recipient, delete this message.  If you are not the
> intended recipient, disclosing, copying, distributing, or taking any
> action based on this message is strictly prohibited.
> 
> 
> 
From mp3.aac.mp4  gmail.com  Wed Nov 23 16:19:10 2005
From: mp3.aac.mp4  gmail.com (tech list)
Date: Wed Nov 23 14:44:26 2005
Subject: [Mp4-tech] [System][AAC][3gp]: SBR suooprt in .3gp file format
In-Reply-To: <C6D6960055FE3E4FBE5D4A8D40D8F0760133E284@torcaexmb2.atitech.com>
References: <C6D6960055FE3E4FBE5D4A8D40D8F0760133E284@torcaexmb2.atitech.com>
Message-ID: <409a09b90511230249v63761380w66c1d561c391483d@mail.gmail.com>
Ryan, you probably need not bother, as the SBR implementation would be able
to extract it's bitstream from the AAC fill elements....
On 11/15/05, Ryan Lei <rlei@ati.com> wrote:
>
> Hi,
>
> Is there any document available about how to support AAC SBR in 3gp file
> format? What is the difference in the 'mp4a' box for SBR and AAC baseline?
>
> Thanks
>
> Ryan
>
> _______________________________________________
> NOTE: Please use clear subject lines for your posts. Include [audio,
> [video], [systems], [general] or another apppropriate identifier to indicate
> the type of question you have.
>
> Note: Conduct on the mailing list is subject to the Antitrust guidelines
> found at
> http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051123/054bfc43/attachment.html
From a26242  alunos.det.ua.pt  Thu Nov 24 06:22:25 2005
From: a26242  alunos.det.ua.pt (Aires Gouveia)
Date: Thu Nov 24 16:07:01 2005
Subject: [Mp4-tech] ISO Working Group-11 documents REQUEST URGENT
Message-ID: <web-3545810@controller.cgpmail.ua.pt>
Hi,
I'm looking for w3908.doc document from ISO Working 
Group-11 documents that was on the link 
http://mpeg.telecomitalialab.com/working_documents/mpeg-04/visual/video_vm.zip 
about MPEG-4 and no longer works...
Can someone please send it to
a26242@alunos.det.ua.pt ASAP?
Tks & Rgds,
Aires Gouveia
From devoo_v  yahoo.co.in  Fri Nov 25 13:10:34 2005
From: devoo_v  yahoo.co.in (devi vijayan)
Date: Sun Nov 27 17:09:04 2005
Subject: [Mp4-tech] request for study material
Message-ID: <20051125131034.57826.qmail@web8415.mail.in.yahoo.com>
  Sir
  I am doing my post graduation in India. I have to take a class seminar regarding ?sprite coding?. Sir can you send basic materials regarding this
  tks&rgds
  devi
---------------------------------
 Enjoy this Diwali with Y! India Click here
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051125/21a9d538/attachment-0001.html
From Stephane.Pechard  univ-nantes.fr  Fri Nov 25 19:08:29 2005
From: Stephane.Pechard  univ-nantes.fr (=?iso-8859-1?Q?St=E9phane_PECHARD?=)
Date: Sun Nov 27 17:16:40 2005
Subject: [Mp4-tech] [H.264] strange rates
Message-ID: <57606.193.52.89.12.1132942109.squirrel@webmail.univ-nantes.fr>
Hi all,
Again with a H.264 reference codec problem using main profile. We have the
GOP structure we want, but the first B frames statistics are strange.
Actually, when we code at 6Mbps, the first B frame is shown to be around
6Mb-sized ! Furthermore, the QP for the two first B frame is 12, but we
specified the QPBSlice to be 30 (we tried 20 and 40 and we still have 12
for the two first B). We have to wait for the whole GOP ended to have some
picture/bit and QP quite normal.
We couldn't find any parameter in the configuration that could explain or
modify this result, except when RateControlEnable is put to 0 (that we
don't want of course).
Does anyone know where this problem could come from ? Which parameter
could be able to modify this ? We tried to modify BList0References,
BList1References and RDOptimization.
Thanx in advance.
--
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
From mp4-tech-owner  lists.mpegif.org  Sun Nov 27 21:19:05 2005
From: mp4-tech-owner  lists.mpegif.org (MPEGIF List Bounces)
Date: Sun Nov 27 21:33:55 2005
Subject: FW: [Mp4-tech] [H.264] strange rates
Message-ID: <000901c5f38f$d10229e0$0200a8c0@corp.intertrust.com>
 forwarding bounced message
-----Original Message-----
From: Tourapis, Alexis [mailto:Alexis.Tourapis@dolby.net] 
Sent: Sunday, 27 November 2023 20:14
To: St?phane PECHARD; mp4-tech@lists.mpegif.org
Subject: RE: [Mp4-tech] [H.264] strange rates
Dear St?phane,
The RC of the reference software is not a very good RC , quite problematic,
and does not make proper adjustments for B slices. It also has no support of
referenced B coded pictures, or hierarchical encoding. In any case,
parameters which affect the QP are not the ones you are changing (they only
affect number of references and RDO usage) are "QPBSlice",
"BRefPicQPOffset", "PyramidLevelQPEnable", and "ExplicitPyramidFormat". The
last three however are considered only if RC is disabled and if you use
Referenced B coded pictures, while the last two also make sense during
hierarchical coding. 
Maybe try modifying the parameters "InitialQP" and "BasicUnit" as well.
Maybe a simplistic selection of BasicUnit=TotalMBsInPicture would be a good
start. Btw note that the RC does not also support interlace properly and
performance loss of even 3-4dB has been reported with weird QP
flunctuations. 
Best Regards,
Alexis
	-----Original Message----- 
	From: mp4-tech-bounces@lists.mpegif.org on behalf of St?phane
PECHARD 
	Sent: Fri 11/25/2005 10:08 AM 
	To: mp4-tech@lists.mpegif.org 
	Cc: 
	Subject: [Mp4-tech] [H.264] strange rates
	Hi all,
	Again with a H.264 reference codec problem using main profile. We
have the
	GOP structure we want, but the first B frames statistics are
strange.
	Actually, when we code at 6Mbps, the first B frame is shown to be
around
	6Mb-sized ! Furthermore, the QP for the two first B frame is 12, but
we
	specified the QPBSlice to be 30 (we tried 20 and 40 and we still
have 12
	for the two first B). We have to wait for the whole GOP ended to
have some
	picture/bit and QP quite normal.
	We couldn't find any parameter in the configuration that could
explain or
	modify this result, except when RateControlEnable is put to 0 (that
we
	don't want of course).
	Does anyone know where this problem could come from ? Which
parameter
	could be able to modify this ? We tried to modify BList0References,
	BList1References and RDOptimization.
	Thanx in advance.
	--
	~ st?phane p?chard
	~ phd student | irccyn-ivc | france
	~ http://www.irccyn.ec-nantes.fr/~pechard
	~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
	_______________________________________________
	NOTE: Please use clear subject lines for your posts. Include [audio,
[video], [systems], [general] or another apppropriate identifier to indicate
the type of question you have.
	Note: Conduct on the mailing list is subject to the Antitrust
guidelines found at
http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-----------------------------------------
This message (including any attachments) may contain confidential
information intended for a specific individual and purpose.  If you are
not the intended recipient, delete this message.  If you are not the
intended recipient, disclosing, copying, distributing, or taking any
action based on this message is strictly prohibited.
From Stephane.Pechard  univ-nantes.fr  Mon Nov 28 15:16:30 2005
From: Stephane.Pechard  univ-nantes.fr (=?iso-8859-1?Q?St=E9phane_PECHARD?=)
Date: Tue Nov 29 15:36:49 2005
Subject: [Mp4-tech] [H.264] strange rates
In-Reply-To: <7272EE229DA1AA48B47EBDC47EB0C68B019D919A@sapphire.dolby.net>
References: <7272EE229DA1AA48B47EBDC47EB0C68B019D919A@sapphire.dolby.net>
Message-ID: <58441.193.52.89.12.1133187390.squirrel@webmail.univ-nantes.fr>
> Maybe a simplistic selection of BasicUnit=TotalMBsInPicture would be a
> good start.
Thanx for this, I have "more logical" QP now. But the bitrate is still way
too much in the first GOP. Here is the result of a sequence coded at
4Mbps, the first frame (an IDR) is more than 4Mb in itself !
Is RC that bad in the codec ?
Following I put the Rate control part of the codec configuration file.
Do you see where it can come from ?
Best Regards,
St?phane
-------------------------------------------------------------------------------
  Frame  Bit/pic    QP   SnrY    SnrU    SnrV    Time(ms) MET(ms) Frm/Fld Ref
-------------------------------------------------------------------------------
0000(NVB)     184
0000(IDR) 4646640   24  38.746  41.417  42.346    262750       0    FRM    1
0002(P)   2589144   24  36.641  40.568  41.922    406813  133127    FLD    1
0001(B)   2315984   23  37.184  40.953  41.997    678656  360966    FRM    0
0004(P)   2420272   24  36.678  40.594  42.005    502485  230955    FLD    1
0003(B)   1302240   25  35.453  40.652  41.858    646719  361664    FRM    0
0006(P)   1725304   26  35.331  40.291  41.642    475844  231024    FLD    1
0005(B)    730480   27  34.186  40.523  41.599    614188  359320    FRM    0
0008(P)   1151880   28  33.994  39.910  41.242    458672  234749    FLD    1
0007(B)    477096   29  33.395  40.107  41.233    594360  359971    FRM    0
0010(P)    854824   30  32.636  39.578  41.016    443391  232447    FLD    1
0009(B)    369560   31  32.291  39.709  41.065    582438  360746    FRM    0
0012(I)   3365712   26  37.142  40.990  42.101    231515       0    FRM    1
0011(B)    386008   31  32.319  39.203  40.984    584781  361285    FLD    0
0014(P)   1643480   26  35.257  39.673  41.389    478859  231344    FLD    1
0013(B)    336888   32  31.423  39.605  40.850    580032  361981    FLD    0
0016(P)   1800504   26  35.246  39.889  41.379    481860  232067    FLD    1
0015(B)    536848   30  32.165  38.836  40.722    596828  362036    FRM    0
0018(P)   1309904   28  33.833  39.232  40.895    463250  230504    FLD    1
0017(B)    550616   29  32.993  39.028  40.982    598781  361426    FRM    0
0020(P)    873656   30  32.530  38.846  40.851    444516  231419    FLD    1
0019(B)    441440   31  31.653  39.010  40.790    586875  360287    FRM    0
0022(P)    744536   32  30.882  38.077  40.198    434516  228781    FLD    1
0021(B)    282840   33  30.171  38.084  40.243    573640  360156    FRM    0
0024(I)   3035440   28  35.436  39.488  41.187    220172       0    FRM    1
0023(B)    265912   33  30.757  38.439  40.490    572141  361016    FRM    0
0026(P)   1250688   28  33.807  38.976  40.859    461765  231844    FLD    1
0025(B)    164880   34  30.652  38.570  40.644    563922  359951    FLD    0
########################################################################################
#Rate control
########################################################################################
RateControlEnable    =      1   # 0 Disable, 1 Enable
Bitrate              =  4000000   # Bitrate(bps)
InitialQP            =     24   # Initial Quantization Parameter for the
first I frame
                                # InitialQp depends on two values: Bits
Per Picture,
                                # and the GOP length
BasicUnit            =     8160   # Number of MBs in the basic unit
                                # should be a fractor of the total number
                                # of MBs in a frame
ChannelType          =      0   # type of channel( 1=time varying channel;
0=Constant channel)
--
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
From girish  dgbmicro.com  Mon Nov 28 19:51:35 2005
From: girish  dgbmicro.com (Girish Shenoy)
Date: Tue Nov 29 15:36:57 2005
Subject: [Mp4-tech] mp4v content in 3gp files
Message-ID: <438B126F.4060202@dgbmicro.com>
Hi All,
First, I note that Quicktime (and also RealPlayer) plays .mp4 files (i 
played some mp4 files successfully).
Also both these are able to play .3gp files with non-mp4 content (I 
played some files successfully here too).
In fact the 3gp file format uses the format defined my mp4 to store the 
mpeg-4 content.
I noticed that I am not able to play 3gp files with mp4v content using 
either of the players. Quicktime gives me an error -2010 "movie contains 
some invalid data" (quicktime player 7.0.3).
Certain other open source (linux based) players are however able to play 
the same files. Theoretically any player able to play 3gp file and mp4 
files should be able to play mp4v content from a 3gp file.Since both 3gp 
and mp4 files are essentially the same base format, I examined the mp4 
file that was played successfully. The only difference I found was that 
this mp4 file had an additional "iods" box/atom.
But the "iods" box/atom is not a mandatory box and in fact the 3gpp file 
format spec mentions that these mpeg-4 architectural elements need not 
be present and shall be ignored.
Am I wrong in saying that iods is not required, is this the problem?
Has anyone else faced this problem? Anybody know any answers?
I would be grateful to any information on the above.
Thanks in advance,
Girish
From Alexis.Tourapis  dolby.net  Mon Nov 28 09:50:57 2005
From: Alexis.Tourapis  dolby.net (Tourapis, Alexis)
Date: Tue Nov 29 15:37:05 2005
Subject: [Mp4-tech] [H.264] strange rates
Message-ID: <7272EE229DA1AA48B47EBDC47EB0C68B01A4DE4E@sapphire.dolby.net>
Dear St?phane,
The authors claim that the use of InitialQP=0 tries to perform an estimate for the QP value that should be used for I slices. That may perform a bit better I guess under the best situations than the use of an explicit value (24 in your case). 
In general, and if your requirements permit this, you can get a rather considerably (without exaggeration) performance by using a certain type of lookahead or even two/multi-pass encoding strategies (they do not have to be that complex but that again is part of your requirements). Unfortunately no lookahead is used by this scheme, while as I have said, the scheme was never designed with interlace in mind (statistics are not exactly correctly collected and used in the software). 
Best regards,
Alexis
-----Original Message-----
From: St?phane PECHARD [mailto:Stephane.Pechard@univ-nantes.fr] 
Sent: Monday, November 28, 2023 6:17 AM
To: Tourapis, Alexis
Cc: mp4-tech@lists.mpegif.org
Subject: RE: [Mp4-tech] [H.264] strange rates
> Maybe a simplistic selection of BasicUnit=TotalMBsInPicture would be a 
> good start.
Thanx for this, I have "more logical" QP now. But the bitrate is still way too much in the first GOP. Here is the result of a sequence coded at 4Mbps, the first frame (an IDR) is more than 4Mb in itself !
Is RC that bad in the codec ?
Following I put the Rate control part of the codec configuration file.
Do you see where it can come from ?
Best Regards,
St?phane
-------------------------------------------------------------------------------
  Frame  Bit/pic    QP   SnrY    SnrU    SnrV    Time(ms) MET(ms) Frm/Fld Ref
-------------------------------------------------------------------------------
0000(NVB)     184
0000(IDR) 4646640   24  38.746  41.417  42.346    262750       0    FRM    1
0002(P)   2589144   24  36.641  40.568  41.922    406813  133127    FLD    1
0001(B)   2315984   23  37.184  40.953  41.997    678656  360966    FRM    0
0004(P)   2420272   24  36.678  40.594  42.005    502485  230955    FLD    1
0003(B)   1302240   25  35.453  40.652  41.858    646719  361664    FRM    0
0006(P)   1725304   26  35.331  40.291  41.642    475844  231024    FLD    1
0005(B)    730480   27  34.186  40.523  41.599    614188  359320    FRM    0
0008(P)   1151880   28  33.994  39.910  41.242    458672  234749    FLD    1
0007(B)    477096   29  33.395  40.107  41.233    594360  359971    FRM    0
0010(P)    854824   30  32.636  39.578  41.016    443391  232447    FLD    1
0009(B)    369560   31  32.291  39.709  41.065    582438  360746    FRM    0
0012(I)   3365712   26  37.142  40.990  42.101    231515       0    FRM    1
0011(B)    386008   31  32.319  39.203  40.984    584781  361285    FLD    0
0014(P)   1643480   26  35.257  39.673  41.389    478859  231344    FLD    1
0013(B)    336888   32  31.423  39.605  40.850    580032  361981    FLD    0
0016(P)   1800504   26  35.246  39.889  41.379    481860  232067    FLD    1
0015(B)    536848   30  32.165  38.836  40.722    596828  362036    FRM    0
0018(P)   1309904   28  33.833  39.232  40.895    463250  230504    FLD    1
0017(B)    550616   29  32.993  39.028  40.982    598781  361426    FRM    0
0020(P)    873656   30  32.530  38.846  40.851    444516  231419    FLD    1
0019(B)    441440   31  31.653  39.010  40.790    586875  360287    FRM    0
0022(P)    744536   32  30.882  38.077  40.198    434516  228781    FLD    1
0021(B)    282840   33  30.171  38.084  40.243    573640  360156    FRM    0
0024(I)   3035440   28  35.436  39.488  41.187    220172       0    FRM    1
0023(B)    265912   33  30.757  38.439  40.490    572141  361016    FRM    0
0026(P)   1250688   28  33.807  38.976  40.859    461765  231844    FLD    1
0025(B)    164880   34  30.652  38.570  40.644    563922  359951    FLD    0
########################################################################################
#Rate control
########################################################################################
RateControlEnable    =      1   # 0 Disable, 1 Enable
Bitrate              =  4000000   # Bitrate(bps)
InitialQP            =     24   # Initial Quantization Parameter for the
first I frame
                                # InitialQp depends on two values: Bits Per Picture,
                                # and the GOP length
BasicUnit            =     8160   # Number of MBs in the basic unit
                                # should be a fractor of the total number
                                # of MBs in a frame
ChannelType          =      0   # type of channel( 1=time varying channel;
0=Constant channel)
--
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
-----------------------------------------
This message (including any attachments) may contain confidential
information intended for a specific individual and purpose.  If you are
not the intended recipient, delete this message.  If you are not the
intended recipient, disclosing, copying, distributing, or taking any
action based on this message is strictly prohibited.
From singer  apple.com  Mon Nov 28 10:00:17 2005
From: singer  apple.com (Dave Singer)
Date: Tue Nov 29 15:37:11 2005
Subject: [Mp4-tech] Re:movie flagment in mp4 file format!
In-Reply-To: <000e01c5ef0e$b3f22a60$1a040180@ccdomain.com>
References: <000e01c5ef0e$b3f22a60$1a040180@ccdomain.com>
Message-ID: <p06230928bfb0f564c99b@[81.254.68.167]>
At 10:44  +0800 22/11/05, ???Y ? wrote:
>?
>
>Dear expert:
>
>Thanks Dave Singer's reply. I was also confused 
>by data-offset field in trun box, is it also 
>like stco box in regular movies? Can you tell me 
>relationship between base-data-offset in tfhd 
>box and data-offset in trun box?
as the spec says, the tfhd data offset sets an 
explicit starting offset for the data offsets in 
the truns.  So you add the tfhd data offset to 
the trun data offset.  That makes a value 
logically the same as those in the chunk offset 
box (stco).
>How can stsc box in regular expand from movie frame boxes?
think of each track run as making a new chunk.
>It use base-data-offset to make of his chunk 
>offset, sample count in trun box to make of 
>samples-per-chunk in regular stsc box, 
>entry-count are add one, first-chunk are the 
>last chunk in moov added one , and one movie 
>fragment make up of one chunk, that is 
>specification design philosophy,my understand is 
>right ?
>
>Thanks very much
>
>Xiongshu,deng
>
><mailto:Dengxiongshu@yulong.com>Dengxiongshu@yulong.com
>
>dengxiongshu@hotmail.com
-- 
David Singer
Apple Computer/QuickTime
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051128/6f0749bc/attachment.html
From Stephane.Pechard  univ-nantes.fr  Tue Nov 29 14:06:38 2005
From: Stephane.Pechard  univ-nantes.fr (=?iso-8859-1?Q?St=E9phane_PECHARD?=)
Date: Tue Nov 29 15:37:18 2005
Subject: [Mp4-tech] [H.264] strange rates
In-Reply-To: <7272EE229DA1AA48B47EBDC47EB0C68B01A4DE4E@sapphire.dolby.net>
References: <7272EE229DA1AA48B47EBDC47EB0C68B01A4DE4E@sapphire.dolby.net>
Message-ID: <50151.193.52.89.12.1133269598.squirrel@webmail.univ-nantes.fr>
Thanks for all these precisions, this is very useful.
I've been said that these strange high rates at the beginning of the
coding are due to the fact that the buffer is infinite. This would explain
that the coder take too much bitrate for the first frames then is more
reasonnable for the following. Am I far from truth ?
We considered to force buffer regulation by adding some frames before the
actual sequence. These frames would be constituted of the first GOP
symetrically put before the first frame.
Another question, during coding of a simple 176*144 YUV sequence, we have
the warning "CABAC stuffing words = 123" (or another value), what does it
mean ? We don't have any of this with our HDTV sequences.
Best regards,
St?phane
--
~ st?phane p?chard
~ phd student | irccyn-ivc | france
~ http://www.irccyn.ec-nantes.fr/~pechard
~~~~~~ pgp : hkp://subkeys.pgp.net ~~~~~~
From qpqiu  trident.com.cn  Wed Nov 30 11:09:50 2005
From: qpqiu  trident.com.cn (Qiu Qiuping)
Date: Wed Nov 30 13:40:43 2005
Subject: [Mp4-tech] (no subject)
Message-ID: <707EF4D95EF97A44B63D407B7953306501AC6780@tmtpms.TMT.COM>
Dear Experts,
  I have a question regarding h.264 decoder implementation. How to output pictures in DPB to display, especially, when there is no SEI information available?
Who can help me, thanks.
?
Qiuping Qiu
From girish  dgbmicro.com  Wed Nov 30 10:57:31 2005
From: girish  dgbmicro.com (Girish Shenoy)
Date: Wed Nov 30 13:40:50 2005
Subject: [Mp4-tech][systems][ISO-fileformat] timescale in mvhd and mdhd
Message-ID: <438D3843.5050908@dgbmicro.com>
Hi All,
In the case where an iso format file has multiple tracks what would be 
the value of timescale field in mvhd?
The following possibilities cross my mind:
1. It has to be the LCM of the timescales in the mdhd box with the 
respective mdhd boxes maintaining their natural timescales.
2.  Use the LCM of timescales of all tracks to populate mvhd and all 
mdhd boxes (ie all the mdhd and mvhd in the file have the same value in 
timescale field; namely the LCM) and the decode/composition deltas for 
each track need to be multiplied with the factor obtained by dividing 
the LCM-Timescale with the track's natural timescale.
3. Mvhd timescale is not too relevant. The mdhd timescales need to 
necessarily match the natural timescales of the corresponding track. 
Timescale in mvhd does not matter.
Which one of the above (or otherwise) is the accepted way to populate 
mvhd and mdhd timescales?
Regards,
Girish
From magnus.hoem  popwire.com  Wed Nov 30 07:34:15 2005
From: magnus.hoem  popwire.com (Magnus Hoem)
Date: Wed Nov 30 13:40:56 2005
Subject: [Mp4-tech] mp4v content in 3gp files
In-Reply-To: <438B126F.4060202@dgbmicro.com>
References: <438B126F.4060202@dgbmicro.com>
Message-ID: <A1CA4001-4BAB-464C-B044-05D2318F90C0@popwire.com>
Hi Girish,
mp4 and 3gpp are basically the same format, but there are minor  
diifferences. In mp4 the iods-box is mandatory (if I remember  
correctly), and it is not in 3gp. It is however not surprising that  
QT can play both (they claim to support both, and can create both  
types correctly). Mp4 supports mpeg4 and H264 video along with AAC  
audio. 3gpp can also contain H263 video and AMR audio, (which is not  
supported in mp4).
About the mp4v-file. Since the release of the video-enabled iPod  
there is some confusion about this extension. It used to be a 14496-2  
(mpeg4) elementary stream, but now Apple uses it for a more or less  
regular mov/3gpp-file with mpeg4 or h264 content (along with AAC  
audio) to be played on the iPod. Which one are you refering to? I  
think QT will have a problem playing the elementary stream, but if  
you insert it correctly into a mov, 3gp or mp4 container it should  
not be a problem at all. The iPod file should of course not cause a  
problem in QT.
Best regards,
Magnus
On 28 nov 2005, at 15.21, Girish Shenoy wrote:
> Hi All,
>
> First, I note that Quicktime (and also RealPlayer) plays .mp4 files  
> (i played some mp4 files successfully).
> Also both these are able to play .3gp files with non-mp4 content (I  
> played some files successfully here too).
> In fact the 3gp file format uses the format defined my mp4 to store  
> the mpeg-4 content.
>
> I noticed that I am not able to play 3gp files with mp4v content  
> using either of the players. Quicktime gives me an error -2010  
> "movie contains some invalid data" (quicktime player 7.0.3).
>
> Certain other open source (linux based) players are however able to  
> play the same files. Theoretically any player able to play 3gp file  
> and mp4 files should be able to play mp4v content from a 3gp  
> file.Since both 3gp and mp4 files are essentially the same base  
> format, I examined the mp4 file that was played successfully. The  
> only difference I found was that this mp4 file had an additional  
> "iods" box/atom.
>
> But the "iods" box/atom is not a mandatory box and in fact the 3gpp  
> file format spec mentions that these mpeg-4 architectural elements  
> need not be present and shall be ignored.
>
> Am I wrong in saying that iods is not required, is this the problem?
> Has anyone else faced this problem? Anybody know any answers?
>
> I would be grateful to any information on the above.
>
> Thanks in advance,
> Girish
> _______________________________________________
> NOTE: Please use clear subject lines for your posts. Include  
> [audio, [video], [systems], [general] or another apppropriate  
> identifier to indicate the type of question you have.
>
> Note: Conduct on the mailing list is subject to the Antitrust  
> guidelines found at http://www.mpegif.org/public/documents/vault/mp- 
> out-30042-Antitrust.php
**********************************************
Popwire Technology
Magnus Hoem
Senior Research Engineer
magnus.hoem@popwire.com
?rsta?ngsv?gen 19 B
Box 47612
SE-117 94 Stockholm, Sweden
Phone: +46 8 579 116 00
Direct: +46 8 579 112 12
Mobile: +46 733 25 44 34
http://www.popwire.com
***********************************************
This message, including any attachments may contain confidential and
privileged material;  it is intended only for the person to whom it is
addressed. Its contents do not constitute a commitment by Popwire except
where provided for in a written and undersigned agreement.
From girish  dgbmicro.com  Wed Nov 30 12:47:02 2005
From: girish  dgbmicro.com (Girish Shenoy)
Date: Wed Nov 30 13:41:03 2005
Subject: [Mp4-tech] mp4v content in 3gp files
In-Reply-To: <A1CA4001-4BAB-464C-B044-05D2318F90C0@popwire.com>
References: <438B126F.4060202@dgbmicro.com>
	<A1CA4001-4BAB-464C-B044-05D2318F90C0@popwire.com>
Message-ID: <438D51EE.2010901@dgbmicro.com>
Thanks Magnus !
As it always happens, one struggles for an answer for days and finally 
decides to write to a forum to get help. Just after that is done, one 
finds his answer.
I did not have any mp4v files. I was having a .3gp file with mp4v 
content and a .mp4 file with mp4v content.
.mp4 plays OK.
.3gp had a problem.
The answer to the problem as I discovered was simple. There seems to be 
a problem with the decoder-specific-info (the concatenated VO headers) 
in the problematic file. Just out of curiosity I just replaced the 
decoder specific info in the problematic file with the same from the 
file which was being successfully played by QT. Although in a strict 
sense, the exercise I did was outrageous, the results were pleasing. 
(Probably because the video in both files had similar properties). And 
lo!! QT started playing the file.
Now I am in the process of analyzing both decoder specific info from 
both files to get to the bottom of it. Once that happens, I guess I will 
update this thread with the root cause (if it is relevant to the members).
Btw!! thanks a lot Magnus for all that information. I was not aware of 
many of the facts you presented.
Thanks,
Girish
Magnus Hoem wrote:
> Hi Girish,
> mp4 and 3gpp are basically the same format, but there are minor  
> diifferences. In mp4 the iods-box is mandatory (if I remember  
> correctly), and it is not in 3gp. It is however not surprising that  
> QT can play both (they claim to support both, and can create both  
> types correctly). Mp4 supports mpeg4 and H264 video along with AAC  
> audio. 3gpp can also contain H263 video and AMR audio, (which is not  
> supported in mp4).
>
> About the mp4v-file. Since the release of the video-enabled iPod  
> there is some confusion about this extension. It used to be a 14496-2  
> (mpeg4) elementary stream, but now Apple uses it for a more or less  
> regular mov/3gpp-file with mpeg4 or h264 content (along with AAC  
> audio) to be played on the iPod. Which one are you refering to? I  
> think QT will have a problem playing the elementary stream, but if  
> you insert it correctly into a mov, 3gp or mp4 container it should  
> not be a problem at all. The iPod file should of course not cause a  
> problem in QT.
>
> Best regards,
> Magnus
>
> On 28 nov 2005, at 15.21, Girish Shenoy wrote:
>
>> Hi All,
>>
>> First, I note that Quicktime (and also RealPlayer) plays .mp4 files  
>> (i played some mp4 files successfully).
>> Also both these are able to play .3gp files with non-mp4 content (I  
>> played some files successfully here too).
>> In fact the 3gp file format uses the format defined my mp4 to store  
>> the mpeg-4 content.
>>
>> I noticed that I am not able to play 3gp files with mp4v content  
>> using either of the players. Quicktime gives me an error -2010  
>> "movie contains some invalid data" (quicktime player 7.0.3).
>>
>> Certain other open source (linux based) players are however able to  
>> play the same files. Theoretically any player able to play 3gp file  
>> and mp4 files should be able to play mp4v content from a 3gp  
>> file.Since both 3gp and mp4 files are essentially the same base  
>> format, I examined the mp4 file that was played successfully. The  
>> only difference I found was that this mp4 file had an additional  
>> "iods" box/atom.
>>
>> But the "iods" box/atom is not a mandatory box and in fact the 3gpp  
>> file format spec mentions that these mpeg-4 architectural elements  
>> need not be present and shall be ignored.
>>
>> Am I wrong in saying that iods is not required, is this the problem?
>> Has anyone else faced this problem? Anybody know any answers?
>>
>> I would be grateful to any information on the above.
>>
>> Thanks in advance,
>> Girish
>> _______________________________________________
>> NOTE: Please use clear subject lines for your posts. Include  [audio, 
>> [video], [systems], [general] or another apppropriate  identifier to 
>> indicate the type of question you have.
>>
>> Note: Conduct on the mailing list is subject to the Antitrust  
>> guidelines found at http://www.mpegif.org/public/documents/vault/mp- 
>> out-30042-Antitrust.php
>
>
> **********************************************
> Popwire Technology
> Magnus Hoem
> Senior Research Engineer
> magnus.hoem@popwire.com
> ?rsta?ngsv?gen 19 B
> Box 47612
> SE-117 94 Stockholm, Sweden
>
> Phone: +46 8 579 116 00
> Direct: +46 8 579 112 12
> Mobile: +46 733 25 44 34
>
> http://www.popwire.com
> ***********************************************
>
> This message, including any attachments may contain confidential and
> privileged material;  it is intended only for the person to whom it is
> addressed. Its contents do not constitute a commitment by Popwire except
> where provided for in a written and undersigned agreement.
>
>
>
>
From praveen.singh  st.com  Wed Nov 30 15:31:35 2005
From: praveen.singh  st.com (Praveen SINGH)
Date: Wed Nov 30 13:41:09 2005
Subject: [Mp4-tech] info/links needed regarding packetizing,
	muxing mpeg4 a/v streams and AVI format
Message-ID: <004801c5f595$0cd2a1f0$1216b40a@dlh.st.com>
Hello,
I have just started with MPEG-4. It would be very kind if someone can send some basic materials and/or links for the below.
1) AVI format.
2) Packetising & Multiplexing mpeg4 video and audio.
Thanks and Best Regards,
Praveen
PS: I have earlier knowledge for mpeg2 pes and muxing.
From gripened  gmail.com  Wed Nov 30 15:42:54 2005
From: gripened  gmail.com (Jayant Chauhan)
Date: Wed Nov 30 13:41:15 2005
Subject: [Mp4-tech] [MP4 Format] Number of chunks in the last run of STSC
	atom
Message-ID: <5b996acb0511300212h5eafdb67u2e24fce1d4a173cf@mail.gmail.com>
Hey guys,
Just wondering, is there an easier and coherent way of finding out the
number of Chunks for the last run of similar chunks in an MP4 file from the
STSC atom?! Because, to find the number of chunks in a particular run, we
need the index for the current and the next run. But for the last run, we
have no further info, so we need to know the number of samples before hand
to be able to do this right ?!? Can somebody please tell me how to do it if
we dont have prior knowledge of the Total Sample numbers in a track?!
regards
Jayant
PS - Btw, does ne one have a MP4 parser which can parse an MP4 file if we
just pass it a fixed size of packets from the file, instead of having the
complete file to parse at a time ?! Am trying to make a filter for the
parser, but the DirectShow requires that I pass the Parser filter a fixed
size from the FileSourceFilter (IAsyncReader)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051130/7ee63b03/attachment.html
From Guido.Franceschini  TILAB.COM  Wed Nov 30 16:36:53 2005
From: Guido.Franceschini  TILAB.COM (Franceschini Guido)
Date: Wed Nov 30 19:30:17 2005
Subject: R: [Mp4-tech] mp4v content in 3gp files
Message-ID: <F52AC9AADF606D46B13F7F8DBB1F1E9D0177A1E3@EXC01A.cselt.it>
Just a hint: check the profile&level, since for 3GP this is more constrained.
Guido
-----Messaggio originale-----
Da: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] Per conto di Girish Shenoy
Inviato: Wednesday, November 30, 2023 8:17 AM
A: Magnus Hoem
Cc: mp4-tech@lists.mpegif.org
Oggetto: Re: [Mp4-tech] mp4v content in 3gp files
Thanks Magnus !
As it always happens, one struggles for an answer for days and finally 
decides to write to a forum to get help. Just after that is done, one 
finds his answer.
I did not have any mp4v files. I was having a .3gp file with mp4v 
content and a .mp4 file with mp4v content.
.mp4 plays OK.
.3gp had a problem.
The answer to the problem as I discovered was simple. There seems to be 
a problem with the decoder-specific-info (the concatenated VO headers) 
in the problematic file. Just out of curiosity I just replaced the 
decoder specific info in the problematic file with the same from the 
file which was being successfully played by QT. Although in a strict 
sense, the exercise I did was outrageous, the results were pleasing. 
(Probably because the video in both files had similar properties). And 
lo!! QT started playing the file.
Now I am in the process of analyzing both decoder specific info from 
both files to get to the bottom of it. Once that happens, I guess I will 
update this thread with the root cause (if it is relevant to the members).
Btw!! thanks a lot Magnus for all that information. I was not aware of 
many of the facts you presented.
Thanks,
Girish
Magnus Hoem wrote:
> Hi Girish,
> mp4 and 3gpp are basically the same format, but there are minor  
> diifferences. In mp4 the iods-box is mandatory (if I remember  
> correctly), and it is not in 3gp. It is however not surprising that  
> QT can play both (they claim to support both, and can create both  
> types correctly). Mp4 supports mpeg4 and H264 video along with AAC  
> audio. 3gpp can also contain H263 video and AMR audio, (which is not  
> supported in mp4).
>
> About the mp4v-file. Since the release of the video-enabled iPod  
> there is some confusion about this extension. It used to be a 14496-2  
> (mpeg4) elementary stream, but now Apple uses it for a more or less  
> regular mov/3gpp-file with mpeg4 or h264 content (along with AAC  
> audio) to be played on the iPod. Which one are you refering to? I  
> think QT will have a problem playing the elementary stream, but if  
> you insert it correctly into a mov, 3gp or mp4 container it should  
> not be a problem at all. The iPod file should of course not cause a  
> problem in QT.
>
> Best regards,
> Magnus
>
> On 28 nov 2005, at 15.21, Girish Shenoy wrote:
>
>> Hi All,
>>
>> First, I note that Quicktime (and also RealPlayer) plays .mp4 files  
>> (i played some mp4 files successfully).
>> Also both these are able to play .3gp files with non-mp4 content (I  
>> played some files successfully here too).
>> In fact the 3gp file format uses the format defined my mp4 to store  
>> the mpeg-4 content.
>>
>> I noticed that I am not able to play 3gp files with mp4v content  
>> using either of the players. Quicktime gives me an error -2010  
>> "movie contains some invalid data" (quicktime player 7.0.3).
>>
>> Certain other open source (linux based) players are however able to  
>> play the same files. Theoretically any player able to play 3gp file  
>> and mp4 files should be able to play mp4v content from a 3gp  
>> file.Since both 3gp and mp4 files are essentially the same base  
>> format, I examined the mp4 file that was played successfully. The  
>> only difference I found was that this mp4 file had an additional  
>> "iods" box/atom.
>>
>> But the "iods" box/atom is not a mandatory box and in fact the 3gpp  
>> file format spec mentions that these mpeg-4 architectural elements  
>> need not be present and shall be ignored.
>>
>> Am I wrong in saying that iods is not required, is this the problem?
>> Has anyone else faced this problem? Anybody know any answers?
>>
>> I would be grateful to any information on the above.
>>
>> Thanks in advance,
>> Girish
>> _______________________________________________
>> NOTE: Please use clear subject lines for your posts. Include  [audio, 
>> [video], [systems], [general] or another apppropriate  identifier to 
>> indicate the type of question you have.
>>
>> Note: Conduct on the mailing list is subject to the Antitrust  
>> guidelines found at http://www.mpegif.org/public/documents/vault/mp- 
>> out-30042-Antitrust.php
>
>
> **********************************************
> Popwire Technology
> Magnus Hoem
> Senior Research Engineer
> magnus.hoem@popwire.com
> ?rsta?ngsv?gen 19 B
> Box 47612
> SE-117 94 Stockholm, Sweden
>
> Phone: +46 8 579 116 00
> Direct: +46 8 579 112 12
> Mobile: +46 733 25 44 34
>
> http://www.popwire.com
> ***********************************************
>
> This message, including any attachments may contain confidential and
> privileged material;  it is intended only for the person to whom it is
> addressed. Its contents do not constitute a commitment by Popwire except
> where provided for in a written and undersigned agreement.
>
>
>
>
_______________________________________________
NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have.
Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
Gruppo Telecom Italia - Direzione e coordinamento di Telecom Italia S.p.A.
====================================================================
CONFIDENTIALITY NOTICE
This message and its attachments are addressed solely to the persons
above and may contain confidential information. If you have received
the message in error, be informed that any use of the content hereof
is prohibited. Please return it immediately to the sender and delete
the message. Should you have any questions, please send an e_mail to 
MailAdmin@tilab.com. Thank you
====================================================================
From dong.tian.2  gmail.com  Wed Nov 30 17:45:38 2005
From: dong.tian.2  gmail.com (Dong Tian)
Date: Wed Nov 30 19:30:23 2005
Subject: [Mp4-tech] (no subject)
In-Reply-To: <707EF4D95EF97A44B63D407B7953306501AC6780@tmtpms.TMT.COM>
References: <707EF4D95EF97A44B63D407B7953306501AC6780@tmtpms.TMT.COM>
Message-ID: <a19456410511300745x558c77ecu897d64994f66b18a@mail.gmail.com>
Hi Qiuping,
I happen to know something related to your question. num_reorder_frames in
VUI tells the decoder the minimum number of pictures should be stored in the
DPB before any output. The bumping process specified in Annex C of the
standard can ensure the correct output order.
BRs,
Dong
On 11/30/05, Qiu Qiuping <qpqiu@trident.com.cn> wrote:
>
> Dear Experts,
>   I have a question regarding h.264 decoder implementation. How to output
> pictures in DPB to display, especially, when there is no SEI information
> available?
> Who can help me, thanks.
>
> Qiuping Qiu
>
> _______________________________________________
> NOTE: Please use clear subject lines for your posts. Include [audio,
> [video], [systems], [general] or another apppropriate identifier to indicate
> the type of question you have.
>
> Note: Conduct on the mailing list is subject to the Antitrust guidelines
> found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20051130/bc9f9783/attachment.html
From singer  apple.com  Wed Nov 30 10:42:21 2005
From: singer  apple.com (Dave Singer)
Date: Wed Nov 30 19:30:30 2005
Subject: [Mp4-tech][systems][ISO-fileformat] timescale in mvhd and mdhd
In-Reply-To: <438D3843.5050908@dgbmicro.com>
References: <438D3843.5050908@dgbmicro.com>
Message-ID: <p0623090ebfb3a288e6fa@[17.202.35.52]>
At 10:57  +0530 30/11/05, Girish Shenoy wrote:
>Hi All,
>
>In the case where an iso format file has multiple tracks what would 
>be the value of timescale field in mvhd?
There is no formal requirement for the timescale.  It should be as 
accurate as the content-creator needed to express the values that are 
in that scale (notably track and movie durations).  You are right 
that it might be prudent to avoid round-off errors (if it matters to 
you), and also if sample-accuracy is needed a high timescale (e.g. 
matching the audio) might be good.  Many people (including QT) use a 
'normal' value for movie timescale (QT uses 600 most of the time), so 
this is closer to point 3.
You are right, the media timescales should be appropriate for the 
media (audio sampling rate, and so on).
>
>The following possibilities cross my mind:
>1. It has to be the LCM of the timescales in the mdhd box with the 
>respective mdhd boxes maintaining their natural timescales.
>2.  Use the LCM of timescales of all tracks to populate mvhd and all 
>mdhd boxes (ie all the mdhd and mvhd in the file have the same value 
>in timescale field; namely the LCM) and the decode/composition 
>deltas for each track need to be multiplied with the factor obtained 
>by dividing the LCM-Timescale with the track's natural timescale.
>3. Mvhd timescale is not too relevant. The mdhd timescales need to 
>necessarily match the natural timescales of the corresponding track. 
>Timescale in mvhd does not matter.
>
>Which one of the above (or otherwise) is the accepted way to 
>populate mvhd and mdhd timescales?
>
>Regards,
>Girish
>_______________________________________________
>NOTE: Please use clear subject lines for your posts. Include [audio, 
>[video], [systems], [general] or another apppropriate identifier to 
>indicate the type of question you have.
>
>Note: Conduct on the mailing list is subject to the Antitrust 
>guidelines found at 
>http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php
-- 
David Singer
Apple Computer/QuickTime