From zhenzhongc hotmail.com Thu Aug 1 00:35:38 2002
From: zhenzhongc hotmail.com (Chen Zhenzhong)
Date: Wed Jul 30 14:08:33 2003
Subject: [M4IF Technotes] Mask pictures (binary or shape images) problem
References: <001401c2389b$21ea8860$66a8758c@GRAPHIC>
Message-ID:
Dear BianSH,
To encode the arbitrarily shaped object, you are required to set proper parameters in the parameter files. In the MS reference software, please set the Alpha.Type=Bianry (or ShapeOnly for encoding the shape only) in the file example.par.
The "seg" file has a format similar to YUV luminance file (no file header). It contains only two values, 0 or 255 where 0 indicate it does not belong to the object and 255 indicate it belongs to the object. You 'd better extract the mask picture frame by frame from the "seg" file before you transform it into bmp format. For example, you may save the extracted picture with the extend name ".raw", which can be opened with PhotoShop such that you may save it as a bmp file.
Regards,
Chen Zhenzhong
----- Original Message -----
From: BianSH
To: technotes@lists.m4if.org
Sent: Wednesday, July 31, 2023 10:04 PM
Subject: [M4IF Technotes] Mask pictures (binary or shape images) problem
I have a question about the Masks for MPEG-4 video sequences on M4IF's resource page.
I download all .seg.zip files and unzip them, but i don't realize how to use the .seg files.
How can I get the mask sequences (ex:bmp files) from the "seg" file?
On the other hand, i wonder that where we can obtain the mask sequences on Internet to test our research?
Grateful for your answers!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020731/06d18044/attachment.html
From alisson dsc.ufpb.br Thu Aug 1 10:29:16 2002
From: alisson dsc.ufpb.br (Alisson Brito)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] Getting frame by frame
Message-ID: <002001c23957$0d36ac20$bf4ba596@alissonpessoal>
Hi,
Is there any enconder that I can ask for it enconde frame by frame, instead of encode all the bitstream?
Thanks,
Alisson
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020801/320b021a/attachment.html
From sg jhu.edu Thu Aug 1 14:24:47 2002
From: sg jhu.edu (Sachin S Gangaputra)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] Quarter pel issues.
Message-ID:
Hi all,
The MPEG4 documentation that I have describes "how" to implement quarter
pel interpolation but fails to describe "why" we choose this 8-tap
symmetric filter. Can someone point me towards references in this
direction that suggest - using this filter is better than the alternative
method of doing a simple linear interpolation.
Further I'm not sure why one needs to perform the mirroring during motion
compensation of a given 8x8 or 16x16 block. In compensating for the next
block from the current block, we do have the whole decoded current frame -
so then do we really need this mirroring - why not just use neighboring
blocks in the interpolation. At the edges of the frame mirroring does make
sense.
Moreover during motion compensation at the encoder mirroring is never
performed as one can interpolate the whole frame in the search of the
right quarter pel motion vector. Shouldn't both sides be consistent in
using or not using mirroring?
Thanks,
Sachin Gangaputra
Graduate Student,
Johns Hopkins Univ.
Baltimore.
From youngall ms8.hinet.net Sat Aug 3 13:58:57 2002
From: youngall ms8.hinet.net (Jacky Young)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] list delete
Message-ID: <009801c23ab5$809168e0$579be33d@laptop>
Please kindly delete my email address from
your mailing address.
My email address is youngall@ms8.hinet.net
Best Regards,
Jacky Young
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020803/e3c668a0/attachment.html
From Christoph.Stadler dynapel.de Mon Aug 5 14:12:40 2002
From: Christoph.Stadler dynapel.de (Christoph Stadler)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] Getting frame by frame
References: <002001c23957$0d36ac20$bf4ba596@alissonpessoal>
Message-ID: <3D4E5DA8.F9F108B@dynapel.de>
> Alisson Brito wrote:
>
> Hi,
> Is there any enconder that I can ask for it enconde frame by frame,
> instead of encode all the bitstream?
> Thanks,
> Alisson
Dear Alisson,
our company (DynaPel Laboratories, Germany) has only recently announced
its new
MPEG-4 Video Encoder/Decoder Software Development Kits
which offer MPEG-4 video in Simple and Advanced Simple Profile and the
very feature you are looking for, i.e. encoding frame by frame.
For a licensing quote please turn to our Sales Team (Sales@DynaPel.com).
Best Regards,
Christoph Stadler
--------------------------------------------------------------
DynaPel Labs. GmbH christoph.stadler@dynapel.de
Dr. Christoph Stadler
Fraunhofer Strasse 9 Tel. +49 (0)89 962428-25
D-85737 Ismaning, GERMANY Fax. +49 (0)89 962428-90
From Michael.Ditze c-lab.de Tue Aug 6 10:40:09 2002
From: Michael.Ditze c-lab.de (Michael Ditze)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] MPEG-4 Questions on Scalability
Message-ID: <3D4F7D59.6020509@c-lab.de>
Hello!
I have some questions with regard to scalability in MPEG-4. Can you
anyone answer some of these questions?
- Is there a difference between temporal and spatial scalability when it
comes to interdependencies bewteen the base stream and the enhancement
stream. I assume that the enhancement stream for temporal scalability
depends on the base stream because of the applied Motion Compensation.
Is that the same for spatial scalability?
- In MPEG-2 you normally have fixed spacing in a Group of Pictures and
between successive I-frames and successive reference frames (I- and
P-frames) that may be represented by the M and N value. Would this be
the same for MPEG-4?
- Even if MPEG-4 is scalable, does that mean that all streams are sent
in any case through the network where the decodes neglects the
enhancement streams if it cannot decode them in time? Or is there a
mechanism that allows to send Enhancement Streams only when the server
knows that the client may decode the stream in time (e.g. though an
Admission Control)
- As the scene description (BIFS) and video streams are sent in
different Elementry streams, is there a mechanism in MPEG-4 that ensures
that the scene description is decoded before the respective streams.
Hoes does that work?
I hope you can answer some of my question as this really important stuff
I need for my thesis. If you have sources where I can find further
information with regard to my questions, I would be very grateful.
Thanks for your efforts anf Kind Regards,
Michael Ditze
--
Michael Ditze
C-LAB, 33094 Paderborn, Germany
F?rstenallee 11, Office FU 304
Voice: +49-5251-60-6124, Fax: +49-5251-60-6065
Email: mailto:michael.ditze@c-lab.de
From stevenp pacificsatellite.com Tue Aug 6 20:09:30 2002
From: stevenp pacificsatellite.com (Steven Pang)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] Microprocessor power
References: <002001c23957$0d36ac20$bf4ba596@alissonpessoal>
Message-ID: <002601c23d39$bd1d4f00$4c5645ca@PACIFIC>
Hi,
Anyone know the Microprocessor processing power requirement to do the MPEG4 decoding and support all the MPEG4 formats? For Digital Satellite / Cable Set Top Box platform?
Best regards,
Steven Pang.
Product Engineer / Technology & MIS Department.
Pacific Satellite International Ltd.
Room 1101 China Merchants Tower,
Shun Tak Centre, 168 Connaught Road Central, Hong Kong
Tel (Direct) : 852 - 2965 3161, Fax: 852 - 2558 0406
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020806/bee30616/attachment.html
From wjf NetworkXXIII.com Tue Aug 6 05:24:47 2002
From: wjf NetworkXXIII.com (William J. Fulco)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] Microprocessor power
In-Reply-To: <002601c23d39$bd1d4f00$4c5645ca@PACIFIC>
Message-ID:
I would be interested in seeing any answers to this question also.
++Bill
William J. Fulco
wjf@NetworkXXIII.com
Logic: When you absolutely, positively
have to refute every fallacy in the room.
-----Original Message-----
From: technotes-admin@lists.m4if.org
[mailto:technotes-admin@lists.m4if.org]On Behalf Of Steven Pang
Sent: Tuesday, August 06, 2023 4:10 AM
To: technotes@lists.m4if.org
Subject: Re: [M4IF Technotes] Microprocessor power
Hi,
Anyone know the Microprocessor processing power requirement to do the
MPEG4 decoding and support all the MPEG4 formats? For Digital Satellite /
Cable Set Top Box platform?
Best regards,
Steven Pang.
Product Engineer / Technology & MIS Department.
Pacific Satellite International Ltd.
Room 1101 China Merchants Tower,
Shun Tak Centre, 168 Connaught Road Central, Hong Kong
Tel (Direct) : 852 - 2965 3161, Fax: 852 - 2558 0406
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020806/09932639/attachment.html
From Michael.Ditze c-lab.de Tue Aug 6 16:46:11 2002
From: Michael.Ditze c-lab.de (Michael Ditze)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] Transmission Order in MPEG-4?
Message-ID: <3D4FD323.5060507@c-lab.de>
Hello,
I have another questions with reagard to the Transmission Order in
MPEG-4. Is it similar to MPEG-2. i.e. if a VoP relies on a succeeding
VoP for decoding, than the latter is sent before. E.g. If a B-VoP relies
on a suceesing P VoP, the P-VoP is inserted in the Transmission Stream
such as it is sent before the B-VoP.
Thanks for your answers.
Cheers, Michael
--
Michael Ditze
C-LAB, 33094 Paderborn, Germany
F?rstenallee 11, Office FU 304
Voice: +49-5251-60-6124, Fax: +49-5251-60-6065
Email: mailto:michael.ditze@c-lab.de
From chl math.uni-bonn.de Wed Aug 7 11:49:35 2002
From: chl math.uni-bonn.de (Christoph Lampert)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] confusion about modulo_time_base, vop_time_increment, ...
Message-ID:
Hi,
I still have problem understanding the correct specifications
for writing timecodes to the VOP header.
First there is modulo_time_base, it consists of a number of
1s followed by a 0, one 1 for every second that passed since the last
GOV header or last module_time_base of the previously decoded I-, S- or
P-VOP.
Am I right, that this does not mean elapsed "full seconds", but is some
kind of "carry" flag, when the ticks-counter "overflows"?
So if we talked minutes and hours, it would be 1 when the clock jumps from
0:59 to 1:00, right?
In real life, the chances of this value ever getting 2 is very small,
correct?
And then, there is vop_time_increment:
It has a value in [0,vop_time_increment_resolution) and the last comment
on it is "The local time base in the units of seconds is recovered by
dividing this value by the vop_time_increment_resolution."
That's simply wrong, isn't it? You will always end up with 0 when dividing
this value by vop_time_increment_resolution... but what _is_ really meant?
Thank you for your help,
Christoph
--
Christoph H. Lampert chl@math,uni-bonn,de | Diese Signature wurde maschi-
Beringstr. 6, Raum 14 Tel. (0228) 73-2948 | nell erstellt und bedarf
Sprechstunden: keine, aber meistens da | keiner Unterschrift. AZ 27B-6
From chl math.uni-bonn.de Wed Aug 7 12:30:07 2002
From: chl math.uni-bonn.de (Christoph Lampert)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] confusion about modulo_time_base, vop_time_increment,
...
In-Reply-To:
Message-ID:
On Wed, 7 Aug 2002, Christoph Lampert wrote:
> Hi,
>
> I still have problem understanding the correct specifications
> for writing timecodes to the VOP header.
>
> First there is modulo_time_base, it consists of a number of
> 1s followed by a 0, one 1 for every second that passed since the last
> GOV header or last module_time_base of the previously decoded I-, S- or
> P-VOP.
sorry, a correction. It was supposed to be "previously displayed I,S,P",
not "decoded".
Christoph
--
Christoph H. Lampert chl@math,uni-bonn,de | Diese Signature wurde maschi-
Beringstr. 6, Raum 14 Tel. (0228) 73-2948 | nell erstellt und bedarf
Sprechstunden: keine, aber meistens da | keiner Unterschrift. AZ 27B-6
From Michael.Ditze c-lab.de Wed Aug 7 14:21:05 2002
From: Michael.Ditze c-lab.de (Michael Ditze)
Date: Wed Jul 30 14:08:34 2003
Subject: [M4IF Technotes] free VOP header fields for additional information?
Message-ID: <3D5102A1.5000506@c-lab.de>
Hello there,
Once again a question concerning the header of a Video Object Plane. Are
there any fields that are free for editing to put in additional
information, e.g. Worst Case Execution time information for the decoder?
Can you name these fields and how long they are?
Regards,
Michael Ditze
--
Michael Ditze
C-LAB, 33094 Paderborn, Germany
F?rstenallee 11, Office FU 304
Voice: +49-5251-60-6124, Fax: +49-5251-60-6065
Email: mailto:michael.ditze@c-lab.de
From arcin atsana.com Wed Aug 7 09:50:06 2002
From: arcin atsana.com (Arcin Bozkurt)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] confusion about modulo_time_base, vop_time_increment, ...
In-Reply-To:
Message-ID: <004701c23e10$f4e71920$d505000a@lumictech.com>
I will be answering your questions in a different order and hopefully will
be clear. I will write my interpretation and how I am doing it.
You are right, modulo_time_base in normal circumstances will never be two.
You can create a setup where it will be, but it would not be very realistic
I believe.
The only time modulo_time_base is non-zero, is when the real time completes
a one full second from the previous time a modulo_time_base was non-zero.
Assume the simple fixed rate case: 30 fps. And the
vop_time_increment_resolution is 300. 300 means each tick is 1/300 seconds
long and the vop_time_increment is given as number of ticks in this
resolution.
Your display times will be 0, 1/30, 2/30, 3/30, .... 29/30, 1, 31/30, etc.
In this case the values for modulo_time_base will be
0, 0, (repeat 29 times), 1, 0 repeat 30 times), 1, 0 (repeat 30 tiems)
and the Vop_time_increment will be
0 10, 20, ... 290, 0 (this corresponds to a "1" in modulo_time_base), 10,
20, ... 290, 0 , etc...
The confusing part is "local time base" related parts I believe. The
standard says:
modulo_time_base: This value represents the local time base in one second
resolution units (1000 milliseconds).
vop_time_increment: The local time base in the units of seconds is recovered
by dividing this value by the vop_time_increment_resolution.
Therefore, local time base in resolution of seconds, is given by
modulo_time_base (as an increment from the previous) and local time base in
resolution of "1/vop_time_increment_resolution" is given by
vop_time_increment. (as an increment from the last modulo_time_base)
If you look at this "division" as a floating point operation you will get
(increment/increment_resoultion) (in seconds) which is exactly what we want.
I hope this clarifies a bit.
Once again, this is my understanding.
-arcin
-----Original Message-----
From: technotes-admin@lists.m4if.org
[mailto:technotes-admin@lists.m4if.org]On Behalf Of Christoph Lampert
Sent: Wednesday, August 07, 2023 4:50 AM
To: technotes@lists.m4if.org
Subject: [M4IF Technotes] confusion about modulo_time_base,
vop_time_increment, ...
Hi,
I still have problem understanding the correct specifications
for writing timecodes to the VOP header.
First there is modulo_time_base, it consists of a number of
1s followed by a 0, one 1 for every second that passed since the last
GOV header or last module_time_base of the previously decoded I-, S- or
P-VOP.
Am I right, that this does not mean elapsed "full seconds", but is some
kind of "carry" flag, when the ticks-counter "overflows"?
So if we talked minutes and hours, it would be 1 when the clock jumps from
0:59 to 1:00, right?
In real life, the chances of this value ever getting 2 is very small,
correct?
And then, there is vop_time_increment:
It has a value in [0,vop_time_increment_resolution) and the last comment
on it is "The local time base in the units of seconds is recovered by
dividing this value by the vop_time_increment_resolution."
That's simply wrong, isn't it? You will always end up with 0 when dividing
this value by vop_time_increment_resolution... but what _is_ really meant?
Thank you for your help,
Christoph
--
Christoph H. Lampert chl@math,uni-bonn,de | Diese Signature wurde maschi-
Beringstr. 6, Raum 14 Tel. (0228) 73-2948 | nell erstellt und bedarf
Sprechstunden: keine, aber meistens da | keiner Unterschrift. AZ 27B-6
_______________________________________________
Technotes mailing list
Technotes@lists.m4if.org
http://lists.m4if.org/mailman/listinfo/technotes
From chl math.uni-bonn.de Wed Aug 7 17:52:41 2002
From: chl math.uni-bonn.de (Christoph Lampert)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] confusion about modulo_time_base, vop_time_increment,
...
In-Reply-To: <004701c23e10$f4e71920$d505000a@lumictech.com>
Message-ID:
Hello,
thank you for your answer. I have one small additional clarification to
ask for:
On Wed, 7 Aug 2002, Arcin Bozkurt wrote:
> I will be answering your questions in a different order and hopefully will
> be clear. I will write my interpretation and how I am doing it.
>
> You are right, modulo_time_base in normal circumstances will never be two.
> You can create a setup where it will be, but it would not be very realistic
> I believe.
>
> The only time modulo_time_base is non-zero, is when the real time completes
> a one full second from the previous time a modulo_time_base was non-zero.
>
> Assume the simple fixed rate case: 30 fps. And the
> vop_time_increment_resolution is 300. 300 means each tick is 1/300 seconds
> long and the vop_time_increment is given as number of ticks in this
> resolution.
>
> Your display times will be 0, 1/30, 2/30, 3/30, .... 29/30, 1, 31/30, etc.
> In this case the values for modulo_time_base will be
> 0, 0, (repeat 29 times), 1, 0 repeat 30 times), 1, 0 (repeat 30 tiems)
> and the Vop_time_increment will be
> 0 10, 20, ... 290, 0 (this corresponds to a "1" in modulo_time_base), 10,
> 20, ... 290, 0 , etc...
That is what I expected, fine. However, this is just for I/P/S-VOPs, not
for B-VOPs. Consider almost a sequence encoded with IBBPBBPBBP...
and only 12fps to keep the lines shorter (vop_time_incr_res=120),
I understand it would be (display order):
I B B P B B P B B P B B P B B P
0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
000 010 020 030 040 050 060 070 080 090 100 110 000 010 020 030
now with 11fps (vop_time_incr_res=110) things get complicated.
Would it be:
I B B P B B P B B P B B P B B P
0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0
000 010 020 030 040 050 060 070 080 090 100 000 010 020 030 040
because reference is always the last I/P/S-VOP, not a B-VOP?
Then I would finally start getting the point...
> The confusing part is "local time base" related parts I believe. The
> standard says:
> modulo_time_base: This value represents the local time base in one second
> resolution units (1000 milliseconds).
> vop_time_increment: The local time base in the units of seconds is recovered
> by dividing this value by the vop_time_increment_resolution.
>
> Therefore, local time base in resolution of seconds, is given by
> modulo_time_base (as an increment from the previous) and local time base in
> resolution of "1/vop_time_increment_resolution" is given by
> vop_time_increment. (as an increment from the last modulo_time_base)
>
> If you look at this "division" as a floating point operation you will get
> (increment/increment_resoultion) (in seconds) which is exactly what we want.
Of course, I didn't think about floating point calculation with integer
variables. Okay, you are right, then. So it's rather a statement
about time "units" than something deep about values...
Yours,
Christoph
> Hi,
>
> I still have problem understanding the correct specifications
> for writing timecodes to the VOP header.
>
> First there is modulo_time_base, it consists of a number of
> 1s followed by a 0, one 1 for every second that passed since the last
> GOV header or last module_time_base of the previously decoded I-, S- or
> P-VOP.
>
> Am I right, that this does not mean elapsed "full seconds", but is some
> kind of "carry" flag, when the ticks-counter "overflows"?
> So if we talked minutes and hours, it would be 1 when the clock jumps from
> 0:59 to 1:00, right?
> In real life, the chances of this value ever getting 2 is very small,
> correct?
>
> And then, there is vop_time_increment:
>
> It has a value in [0,vop_time_increment_resolution) and the last comment
> on it is "The local time base in the units of seconds is recovered by
> dividing this value by the vop_time_increment_resolution."
>
> That's simply wrong, isn't it? You will always end up with 0 when dividing
> this value by vop_time_increment_resolution... but what _is_ really meant?
>
> Thank you for your help,
>
> Christoph
>
> --
> Christoph H. Lampert chl@math,uni-bonn,de | Diese Signature wurde maschi-
> Beringstr. 6, Raum 14 Tel. (0228) 73-2948 | nell erstellt und bedarf
> Sprechstunden: keine, aber meistens da | keiner Unterschrift. AZ 27B-6
>
> _______________________________________________
> Technotes mailing list
> Technotes@lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>
>
--
Christoph H. Lampert chl@math,uni-bonn,de | Diese Signature wurde maschi-
Beringstr. 6, Raum 14 Tel. (0228) 73-2948 | nell erstellt und bedarf
Sprechstunden: keine, aber meistens da | keiner Unterschrift. AZ 27B-6
From sachin smartyantra.com Thu Aug 8 15:53:47 2002
From: sachin smartyantra.com (Sachin Patil)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] Problem : Half Pixel ME
Message-ID: <5.1.0.14.0.20020808143720.00a0edc0@192.168.1.3>
Hello all,
I have an integer pel search Motion Estimation algorithm
implemented for my MPEG4 video encoder. Now I am trying to refine the
search using half pixel resolution. I find that there is an increase in
file size with half pel refined search.
Can anyone tell me the advantages of using Half Pel search with respect to
the following --
1. I want to know if there can be an increase in the output file size.
2. Is half pel refinement used only for an improvement in quality ( quality
measure used : PSNR ) ?
3. What is the increased computational complexity?
Thankyou,
Sachin.
--
Sachin Patil.
Smart Yantra Technologies Pvt. Ltd.
Bangalore
INDIA
Phone : +91 80 363 4917, 363 4918
Email : sachin@smartyantra.com
Website : www.smartyantra.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020808/8e897186/attachment.html
From salo rucus.ru.ac.za Thu Aug 8 14:07:59 2002
From: salo rucus.ru.ac.za (Nik)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] Interactive within interactive
Message-ID: <001b01c23ecb$dc29df10$9679e792@beast>
Hi there,
I understand the concept of abstracting out audio video objects, which
can then be composed as you like, but could one of these objects in turn
be *another* mpeg-4 file with its own interactivity etc.?
So if we had a complex mpeg-4 file exhibiting interactivity (still not
sure-is this likely to be advanced simple profile?), could we then
interact with a similar file (same profile) embedded somewhere within
it?
Thanks for your help!
Nik.
From nico.oorts xmt.be Thu Aug 8 15:37:33 2002
From: nico.oorts xmt.be (Nico Oorts)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] multiple audio tracks into an mp4 file
Message-ID: <3D52660D.2080409@xmt.be>
Hi all,
With the mp4tool (JC Dufourd), I try to make a .mp4 file which contains
multiple audiotracks. I make a scene with an audiosource. This
audiosource has an url to the OD which contains the elementary stream.
After some time I replace the audiosource by another which has also an
url to an OD and ES. The result is an .mp4 which contains the tracks (I
can check this with some tools and size), but play only the first track.
Some ideas why this does not work?
Thx
Nico Oorts
Research team
VRT
From Peter.Haighton m4if.org Thu Aug 8 10:28:08 2002
From: Peter.Haighton m4if.org (Peter Haighton)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] Problem : Half Pixel ME
In-Reply-To: <5.1.0.14.0.20020808143720.00a0edc0@192.168.1.3>
Message-ID:
Sachin,
The answers to your questions are definitely dependant on implendation
details. Depending on what the ME finds as it's best match will determine
whether or not the file size will increase. This is the case also for full
pixel ME since the best match that you obtain through for example SAD of
macroblocks may not translate to the smallest VLC codes.
Half pixel ME is designed to improve quality of the image but again that is
dependant on many factors. For example, if you find that your ME technique
consistently uses many more bits than full pixel ME, and you are compressing
at a fix rate, you may find that the quality decreases. On the otherhand it
is typical that higher quality is obtained both in PSNR and visually.
Increased Complexity comes right down to a case by case basis on how the ME
is built. There are a few ways of building in half pixel ME with very
different complexities.
Peter
--
Peter Haighton
VideoSpheres Inc.
84 Hines Road
Kanata, Ontario
Canada, K2K 3G3
Tel: (613) 270-9646 x3022
Fax: (613) 271-9442
email: peterh@VideoSpheres.com
See http://www.m4if.org for the latest on MPEG-4
-----Original Message-----
From: technotes-admin@lists.m4if.org
[mailto:technotes-admin@lists.m4if.org]On Behalf Of Sachin Patil
Sent: Thursday, August 08, 2023 5:24 AM
To: technotes@lists.m4if.org
Subject: [M4IF Technotes] Problem : Half Pixel ME
Hello all,
I have an integer pel search Motion Estimation algorithm implemented
for my MPEG4 video encoder. Now I am trying to refine the search using half
pixel resolution. I find that there is an increase in file size with half
pel refined search.
Can anyone tell me the advantages of using Half Pel search with respect to
the following --
1. I want to know if there can be an increase in the output file size.
2. Is half pel refinement used only for an improvement in quality ( quality
measure used : PSNR ) ?
3. What is the increased computational complexity?
Thankyou,
Sachin.
--
Sachin Patil.
Smart Yantra Technologies Pvt. Ltd.
Bangalore
INDIA
Phone : +91 80 363 4917, 363 4918
Email : sachin@smartyantra.com
Website : www.smartyantra.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020808/bb5b6cd8/attachment.html
From chl math.uni-bonn.de Thu Aug 8 17:09:47 2002
From: chl math.uni-bonn.de (Christoph Lampert)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] Problem : Half Pixel ME
In-Reply-To: <5.1.0.14.0.20020808143720.00a0edc0@192.168.1.3>
Message-ID:
Hi,
I can answer you something, not as an expert but with some
experience from the Simple Profile Open Source MPEG-4 Codec XviD
(www.xvid.org):
On Thu, 8 Aug 2002, Sachin Patil wrote:
> Hello all,
> I have an integer pel search Motion Estimation algorithm
> implemented for my MPEG4 video encoder. Now I am trying to refine the
> search using half pixel resolution. I find that there is an increase in
> file size with half pel refined search.
>
> Can anyone tell me the advantages of using Half Pel search with respect to
> the following --
> 1. I want to know if there can be an increase in the output file size.
It is (of course) possible that the Half Pel motion vectors need more
space in VLC than the Full Pel ones, and that the gain in prediction
accuracy does not compensate for this. However, if you just
"refine" motion vectors from a Full Pel search by a small amount, I
consider this effect rather unlikely for "normal" input material. You
might want to check if your implementation is correct.
> 2. Is half pel refinement used only for an improvement in quality ( quality
> measure used : PSNR ) ?
For us it has both effects: Smaller bitrate at fixed quantization
parameter, _and_ improvement in quality (more than 1dB gain in PSNR,
often much more).
> 3. What is the increased computational complexity?
That depends only on your implementation.
In our codec on i386 with SIMD optimization, speed with Half Pel usage is
about two thirds of speed without (always remember it's Simple Profile).
This is mostly because we interpolate the whole image, then perform ME
without distriminating between Full Pel and Half Pel positions, and after
that there is a simple +-1 refinement check in x and y of the so far best
motion vector.
Hope that helps,
Christoph
--
Christoph H. Lampert chl@math,uni-bonn,de | Diese Signature wurde maschi-
Beringstr. 6, Raum 14 Tel. (0228) 73-2948 | nell erstellt und bedarf
Sprechstunden: keine, aber meistens da | keiner Unterschrift. AZ 27B-6
From arcin atsana.com Thu Aug 8 12:12:20 2002
From: arcin atsana.com (Arcin Bozkurt)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] confusion about modulo_time_base, vop_time_increment, ...
In-Reply-To:
Message-ID: <003701c23eed$ff530f80$d505000a@lumictech.com>
> -----Original Message-----
> From: technotes-admin@lists.m4if.org
> [mailto:technotes-admin@lists.m4if.org]On Behalf Of Christoph Lampert
> Sent: Wednesday, August 07, 2023 10:53 AM
> To: technotes@lists.m4if.org
> Subject: RE: [M4IF Technotes] confusion about modulo_time_base,
> vop_time_increment, ...
>
>
> Hello,
>
> thank you for your answer. I have one small additional
> clarification to
> ask for:
>
> On Wed, 7 Aug 2002, Arcin Bozkurt wrote:
> > I will be answering your questions in a different order and
> hopefully will
> > be clear. I will write my interpretation and how I am doing it.
> >
> > You are right, modulo_time_base in normal circumstances
> will never be two.
> > You can create a setup where it will be, but it would not
> be very realistic
> > I believe.
> >
> > The only time modulo_time_base is non-zero, is when the
> real time completes
> > a one full second from the previous time a modulo_time_base
> was non-zero.
> >
> > Assume the simple fixed rate case: 30 fps. And the
> > vop_time_increment_resolution is 300. 300 means each tick
> is 1/300 seconds
> > long and the vop_time_increment is given as number of ticks in this
> > resolution.
> >
> > Your display times will be 0, 1/30, 2/30, 3/30, .... 29/30,
> 1, 31/30, etc.
> > In this case the values for modulo_time_base will be
> > 0, 0, (repeat 29 times), 1, 0 repeat 30 times), 1, 0
> (repeat 30 tiems)
> > and the Vop_time_increment will be
> > 0 10, 20, ... 290, 0 (this corresponds to a "1" in
> modulo_time_base), 10,
> > 20, ... 290, 0 , etc...
>
> That is what I expected, fine. However, this is just for
> I/P/S-VOPs, not
> for B-VOPs. Consider almost a sequence encoded with IBBPBBPBBP...
> and only 12fps to keep the lines shorter (vop_time_incr_res=120),
> I understand it would be (display order):
>
> I B B P B B P B B P B B P B B P
> 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0
> 000 010 020 030 040 050 060 070 080 090 100 110 000 010 020 030
>
> now with 11fps (vop_time_incr_res=110) things get complicated.
> Would it be:
>
> I B B P B B P B B P B B P B B P
> 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0
> 000 010 020 030 040 050 060 070 080 090 100 000 010 020 030 040
^^^
The Marked "B" is based on the P at 90. I would say what you wrote is
correct.
for the next one which is a P, it is also based on the same P at 90.
From lcheng62 yahoo.com Sun Aug 11 20:24:13 2002
From: lcheng62 yahoo.com (Liang Cheng)
Date: Wed Jul 30 14:08:35 2003
Subject: [M4IF Technotes] about FGS specification
Message-ID: <20020812022413.6817.qmail@web21106.mail.yahoo.com>
Hi,
Where can I find the latest FGS specification? Is it
in the latest 14496-2?
Thanks,
Liang
__________________________________________________
Do You Yahoo!?
HotJobs - Search Thousands of New Jobs
http://www.hotjobs.com
From rkoenen intertrust.com Sun Aug 11 21:42:34 2002
From: rkoenen intertrust.com (Rob Koenen)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] about FGS specification
Message-ID: <3C124172E7FDD511B510000347426D59011B3727@exchange.epr.com>
FGS is in the following amendment:
ISO/IEC 14496-2:2001/Amd 2:2002 (Streaming video profile)
See
http://www.iso.ch/iso/en/StandardsQueryFormHandler.StandardsQueryFormHandler
?languageCode=en&keyword=&lastSearch=false&isoNumber=14496&isoPartNumber=&IC
S=&stageCode=&stageDate=&committee=ALL&subcommittee=&scope=CATALOGUE&sortOrd
er=ISO
make sure to remove any linebreaks.
Or go to http://www.m4if.org/resources.php and
look under "Obtaining the standard"
Rob
> -----Original Message-----
> From: Liang Cheng [mailto:lcheng62@yahoo.com]
> Sent: Sunday, August 11, 2023 19:24
> To: technotes@lists.m4if.org
> Subject: [M4IF Technotes] about FGS specification
>
>
> Hi,
>
> Where can I find the latest FGS specification? Is it
> in the latest 14496-2?
>
> Thanks,
> Liang
>
> __________________________________________________
> Do You Yahoo!?
> HotJobs - Search Thousands of New Jobs
> http://www.hotjobs.com
> _______________________________________________
> Technotes mailing list
> Technotes@lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>
From salo rucus.ru.ac.za Mon Aug 12 12:42:18 2002
From: salo rucus.ru.ac.za (Nik)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] Live streaming BIFS commands
Message-ID: <003801c241e4$8d4edd20$9679e792@beast>
Hi there. I hope you can help me with this fairly high level question.
In a live mpeg-4 streaming situation, what tool would one use to
manipulate the scene description on the fly so as to remove/add more
live streams etc.?Assume for example, one wished to add a live stream
with a URL that was unknown at the start of transmission.
Do these tools exist, or would they have to be integrated with the
streaming server?
Thanks for your time :)
Nik.
From piccarre elet.polimi.it Wed Aug 14 02:02:18 2002
From: piccarre elet.polimi.it (Luca Piccarreta)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] Wavelet Test Sequences
References: <32808FAD125F8A49A3145D4BB67ADC4024C35D@lanmhs50.rd.francetelecom.fr>
Message-ID: <001901c2431d$79866a80$8d0ed8d4@bigclit>
DMIF and Amendment 7Hi all.
I tried my best but I'm really not able to decoder Envivio wavelet
streams in the sequences on their site.
Can anyone (may be them) tell me how to compose DecoderSpecificInfo
and AU and feed them into the decoder?
This is what I did:
1) Strip the VO_Start code, End codes, etc., just left the VTC (0xbe) start code
2) Write both into a file
3) Decode them with:
-Decoder MS Version 1.0
-Decoder MS Version 2.0 w4711 (tried version 1 and version 2 syntax)
-Decoder MS Version 2.0 w4866 (tried version 1 and version 2 syntax)
Moreover. Anyone knows where the updated test sequences are?
(I mean, the old ms-vfdis-v10-ba1... bs4... vt4, but with the updated syntax
especially for shape and vtc).
Thanks in advance.
Luca Piccarreta
Tilab
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020814/d4130751/attachment.html
From manu.batura patni.com Wed Aug 14 10:23:01 2002
From: manu.batura patni.com (Manu V Batura)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] Distortion Testing for Encoder
Message-ID: <000201c24346$162cc640$5402a8c0@patni.com>
Hi,
As per Part-4 of the standard, it seems that any video encoder that can
generate a compliant bitstream would be
compliant to the standard. However, is there aso a corresponding distortion
measure that needs to be acheived?
(something like RMSE etc). If yes, what are the thresholds that should be
used for testing 'acceptable video quality'/'allowable video distortion'?
Thanks & Regards,
Manu V Batura
From rkoenen intertrust.com Tue Aug 13 23:24:28 2002
From: rkoenen intertrust.com (Rob Koenen)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] Distortion Testing for Encoder
Message-ID: <3C124172E7FDD511B510000347426D59011B377F@exchange.epr.com>
Indeed, any decoder that produces a valid bistream is compliant
to the standard. There are deliberately no prescribed quality
metrics, nor is there any prescribed algorithm to perform the
encoding.
This means there will be good encoders and bad ones...
Best,
Rob
> -----Original Message-----
> From: Manu V Batura [mailto:manu.batura@patni.com]
> Sent: Tuesday, August 13, 2023 20:53
> To: technotes@lists.m4if.org
> Subject: [M4IF Technotes] Distortion Testing for Encoder
>
>
> Hi,
>
> As per Part-4 of the standard, it seems that any video
> encoder that can
> generate a compliant bitstream would be
> compliant to the standard. However, is there aso a
> corresponding distortion
> measure that needs to be acheived?
> (something like RMSE etc). If yes, what are the thresholds
> that should be
> used for testing 'acceptable video quality'/'allowable video
> distortion'?
>
> Thanks & Regards,
> Manu V Batura
>
> _______________________________________________
> Technotes mailing list
> Technotes@lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>
From manu.batura patni.com Wed Aug 14 12:53:38 2002
From: manu.batura patni.com (Manu V Batura)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] Distortion Testing for Encoder
In-Reply-To: <3C124172E7FDD511B510000347426D59011B377F@exchange.epr.com>
Message-ID: <000901c2435b$21029ee0$5402a8c0@patni.com>
Thanks Rob,
But are there any objective tests for distortion that can be used
for testing which encoders are better. If distortion is the competitive
measure, there
should be distortion thresholds available that may be used to call some
encoders good
and some bad.
Thanks & Regards,
Manu
-----Original Message-----
From: technotes-admin@lists.m4if.org
[mailto:technotes-admin@lists.m4if.org]On Behalf Of Rob Koenen
Sent: Wednesday, August 14, 2023 10:54 AM
To: 'manu.batura@patni.com'; technotes@lists.m4if.org
Subject: RE: [M4IF Technotes] Distortion Testing for Encoder
Indeed, any decoder that produces a valid bistream is compliant
to the standard. There are deliberately no prescribed quality
metrics, nor is there any prescribed algorithm to perform the
encoding.
This means there will be good encoders and bad ones...
Best,
Rob
> -----Original Message-----
> From: Manu V Batura [mailto:manu.batura@patni.com]
> Sent: Tuesday, August 13, 2023 20:53
> To: technotes@lists.m4if.org
> Subject: [M4IF Technotes] Distortion Testing for Encoder
>
>
> Hi,
>
> As per Part-4 of the standard, it seems that any video
> encoder that can
> generate a compliant bitstream would be
> compliant to the standard. However, is there aso a
> corresponding distortion
> measure that needs to be acheived?
> (something like RMSE etc). If yes, what are the thresholds
> that should be
> used for testing 'acceptable video quality'/'allowable video
> distortion'?
>
> Thanks & Regards,
> Manu V Batura
>
> _______________________________________________
> Technotes mailing list
> Technotes@lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>
_______________________________________________
Technotes mailing list
Technotes@lists.m4if.org
http://lists.m4if.org/mailman/listinfo/technotes
From rkoenen intertrust.com Wed Aug 14 10:09:55 2002
From: rkoenen intertrust.com (Rob Koenen)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] Distortion Testing for Encoder
Message-ID: <3C124172E7FDD511B510000347426D59011B3788@exchange.epr.com>
> But are there any objective tests for distortion that can be used
> for testing which encoders are better. If distortion is the
> competitive measure, there
> should be distortion thresholds available that may be used to
> call some encoders good and some bad.
The point is that there are no encoding quality requirements in the
MPEG-4 standard. Of course there are many things you can do that go
beyond the standard.
NB - Nothing beats human evaluation, and I have yet to see proof that
anything non-human beats signal-to-noise ratios. (Interestingly, this
point comes up every 2-3 months on this list.)
Best Regards,
Rob
From jp skystream.com Wed Aug 14 15:35:37 2002
From: jp skystream.com (Jeyendran Balakrishnan)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] Distortion Testing for Encoder
Message-ID:
Try the JND Toolkit from David Sarnoff labs
(http://www.sarnoff.com/internet_telecom/jndmetrix/index.asp)
available for about US$ 9K, Windows SW, offline analysis.
Or you can get the more pricey Tektronix PQA box (about $30K
I think, but don't hold me to this), which uses the same
Sarnoff code inside, but (I believe) works real-time and
for streaming content, and probably adds other bells
and whistles.
This generally gives some objective measures of video quality,
taking into account the nature of the human psycho-visual system.
You can use this to compare quality of different encoders
on the same input content.
However, to decide if any one encoder is good enough on
an absolute basis is subjective, and generally
addressable only by having a (sufficiently large)
bunch of people look at the reconstructed output and judge
the quality.
-Jeyendran
-----Original Message-----
From: Manu V Batura [mailto:manu.batura@patni.com]
Sent: Tuesday, August 13, 2023 11:24 PM
To: technotes@lists.m4if.org
Subject: RE: [M4IF Technotes] Distortion Testing for Encoder
Thanks Rob,
But are there any objective tests for distortion that can be used
for testing which encoders are better. If distortion is the competitive
measure, there
should be distortion thresholds available that may be used to call some
encoders good
and some bad.
Thanks & Regards,
Manu
-----Original Message-----
From: technotes-admin@lists.m4if.org
[mailto:technotes-admin@lists.m4if.org]On Behalf Of Rob Koenen
Sent: Wednesday, August 14, 2023 10:54 AM
To: 'manu.batura@patni.com'; technotes@lists.m4if.org
Subject: RE: [M4IF Technotes] Distortion Testing for Encoder
Indeed, any decoder that produces a valid bistream is compliant
to the standard. There are deliberately no prescribed quality
metrics, nor is there any prescribed algorithm to perform the
encoding.
This means there will be good encoders and bad ones...
Best,
Rob
> -----Original Message-----
> From: Manu V Batura [mailto:manu.batura@patni.com]
> Sent: Tuesday, August 13, 2023 20:53
> To: technotes@lists.m4if.org
> Subject: [M4IF Technotes] Distortion Testing for Encoder
>
>
> Hi,
>
> As per Part-4 of the standard, it seems that any video
> encoder that can
> generate a compliant bitstream would be
> compliant to the standard. However, is there aso a
> corresponding distortion
> measure that needs to be acheived?
> (something like RMSE etc). If yes, what are the thresholds
> that should be
> used for testing 'acceptable video quality'/'allowable video
> distortion'?
>
> Thanks & Regards,
> Manu V Batura
>
> _______________________________________________
> Technotes mailing list
> Technotes@lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>
_______________________________________________
Technotes mailing list
Technotes@lists.m4if.org
http://lists.m4if.org/mailman/listinfo/technotes
_______________________________________________
Technotes mailing list
Technotes@lists.m4if.org
http://lists.m4if.org/mailman/listinfo/technotes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020814/52e61bd6/attachment.html
From rkoenen intertrust.com Wed Aug 14 21:36:23 2002
From: rkoenen intertrust.com (Rob Koenen)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] Distortion Testing for Encoder
Message-ID: <3C124172E7FDD511B510000347426D59011B37A9@exchange.epr.com>
Please allow me write a Word of Caution.
While 'objective' measures may be starting to get better than SNR,
and are highly useful for monitoring purposes and optimizing encoder
quality (including making coding decisions), they do not allow one
to choose between one coder and the other.
Their correlation with human subjective quality judgement is still too
low for such purposes.
Kind Regards,
Rob
-----Original Message-----
From: Jeyendran Balakrishnan [mailto:jp@skystream.com]
Sent: Wednesday, August 14, 2023 14:36
To: 'manu.batura@patni.com'; technotes@lists.m4if.org
Subject: RE: [M4IF Technotes] Distortion Testing for Encoder
Try the JND Toolkit from David Sarnoff labs
( http://www.sarnoff.com/internet_telecom/jndmetrix/index.asp
)
available for about US$ 9K, Windows SW, offline analysis.
Or you can get the more pricey Tektronix PQA box (about $30K
I think, but don't hold me to this), which uses the same
Sarnoff code inside, but (I believe) works real-time and
for streaming content, and probably adds other bells
and whistles.
This generally gives some objective measures of video quality,
taking into account the nature of the human psycho-visual system.
You can use this to compare quality of different encoders
on the same input content.
However, to decide if any one encoder is good enough on
an absolute basis is subjective, and generally
addressable only by having a (sufficiently large)
bunch of people look at the reconstructed output and judge
the quality.
-Jeyendran
-----Original Message-----
From: Manu V Batura [ mailto:manu.batura@patni.com
]
Sent: Tuesday, August 13, 2023 11:24 PM
To: technotes@lists.m4if.org
Subject: RE: [M4IF Technotes] Distortion Testing for Encoder
Thanks Rob,
But are there any objective tests for distortion that can be used
for testing which encoders are better. If distortion is the competitive
measure, there
should be distortion thresholds available that may be used to call some
encoders good
and some bad.
Thanks & Regards,
Manu
-----Original Message-----
From: technotes-admin@lists.m4if.org
[ mailto:technotes-admin@lists.m4if.org
]On Behalf Of Rob Koenen
Sent: Wednesday, August 14, 2023 10:54 AM
To: 'manu.batura@patni.com'; technotes@lists.m4if.org
Subject: RE: [M4IF Technotes] Distortion Testing for Encoder
Indeed, any decoder that produces a valid bistream is compliant
to the standard. There are deliberately no prescribed quality
metrics, nor is there any prescribed algorithm to perform the
encoding.
This means there will be good encoders and bad ones...
Best,
Rob
> -----Original Message-----
> From: Manu V Batura [ mailto:manu.batura@patni.com
]
> Sent: Tuesday, August 13, 2023 20:53
> To: technotes@lists.m4if.org
> Subject: [M4IF Technotes] Distortion Testing for Encoder
>
>
> Hi,
>
> As per Part-4 of the standard, it seems that any video
> encoder that can
> generate a compliant bitstream would be
> compliant to the standard. However, is there aso a
> corresponding distortion
> measure that needs to be acheived?
> (something like RMSE etc). If yes, what are the thresholds
> that should be
> used for testing 'acceptable video quality'/'allowable video
> distortion'?
>
> Thanks & Regards,
> Manu V Batura
>
> _______________________________________________
> Technotes mailing list
> Technotes@lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>
_______________________________________________
Technotes mailing list
Technotes@lists.m4if.org
http://lists.m4if.org/mailman/listinfo/technotes
_______________________________________________
Technotes mailing list
Technotes@lists.m4if.org
http://lists.m4if.org/mailman/listinfo/technotes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020814/8275cdd0/attachment.html
From yi.li skyworksinc.com Mon Aug 19 20:18:24 2002
From: yi.li skyworksinc.com (Yi Li)
Date: Wed Jul 30 14:08:36 2003
Subject: [M4IF Technotes] De-blocking/De-ringing
Message-ID: <200208192319.g7JNJYWv017193@mx3.magma.ca>
Hello,
I have some question regarding the de-blocking and de-ringing filters. The
MPEG-4 visual discussed briefly a method to reduce the computational cost of
de-blocking/de-ringing. This is achieved by defining ringing and blocking
semaphores. However, it is not quite clear to me how to use those
semaphores since the descriptions are too short. Does any one have more
detailed info on that? The spec also mentions a reference simply called
M2723 (maybe this is a MPEG-4 spec ref number). Do anyone have a electronic
copy of that document?
Thanks
Yi Li
From rm technopark.lt Wed Aug 21 12:28:04 2002
From: rm technopark.lt (Remigijus)
Date: Wed Jul 30 14:08:37 2003
Subject: [M4IF Technotes] Is standard conform to put mp3 sound in a MPEG-4 stream ?
Message-ID: <001201c248f5$0d693020$3301a8c0@technopark.dom>
Hi All
I would like to ask: Is standard conform to put mp3 sound in a MPEG-4 stream:
I've created mp4 file with mp4creator (MPEG4IP) with mp3 sound track inside. When I'm playing this file with QuickTime 6 It generates Error message:
"File Contains invalid audio data:
An audio Decoder Configuration descriptor corrupted.
Expected ObjectType = Audio(0x40); found object type ^0 instead."
Does it means that QuickTime doesn't supports mp3 sound inside mp4 or something else.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020821/031983b9/attachment.html
From rchandpur tataelxsi.co.in Wed Aug 21 17:47:14 2002
From: rchandpur tataelxsi.co.in (RUTURAJ CHANDPUR )
Date: Wed Jul 30 14:08:37 2003
Subject: [M4IF Technotes] Is standard conform to put mp3 sound in a MPEG-4 stream ?
References: <001201c248f5$0d693020$3301a8c0@technopark.dom>
Message-ID: <014e01c24904$4df4f070$0314010a@tataelxsi.co.in>
Hi,
The objectTypeIndication for mp3 audio is 0x6B as per Table 8 of ISO/IEC 14496-1:2001(E).
Hope this helps u.
rgds,
Ruturaj Anand Chandpur
----- Original Message -----
From: Remigijus
To: technotes@lists.m4if.org
Sent: Wednesday, August 21, 2023 2:58 PM
Subject: [M4IF Technotes] Is standard conform to put mp3 sound in a MPEG-4 stream ?
Hi All
I would like to ask: Is standard conform to put mp3 sound in a MPEG-4 stream:
I've created mp4 file with mp4creator (MPEG4IP) with mp3 sound track inside. When I'm playing this file with QuickTime 6 It generates Error message:
"File Contains invalid audio data:
An audio Decoder Configuration descriptor corrupted.
Expected ObjectType = Audio(0x40); found object type ^0 instead."
Does it means that QuickTime doesn't supports mp3 sound inside mp4 or something else.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020821/04bcd1e0/attachment.html
From rkoenen intertrust.com Wed Aug 21 10:19:44 2002
From: rkoenen intertrust.com (Rob Koenen)
Date: Wed Jul 30 14:08:37 2003
Subject: [M4IF Technotes] Is standard conform to put mp3 sound in a MP
EG-4 stream ?
Message-ID: <3C124172E7FDD511B510000347426D59011B3873@exchange.epr.com>
That's right. In addition, I'd like to note that while mp3 is recognized as
a
native object type at the systems level, there is currently no MPEG-4
Profile that includes mp3 object types, so it may not be supported in
some or even many systems. Popular demand will probably decide where
this will go.
Best Regards,
Rob
-----Original Message-----
From: RUTURAJ CHANDPUR [mailto:rchandpur@tataelxsi.co.in]
Sent: Wednesday, August 21, 2023 4:17
To: Remigijus
Cc: technotes@lists.m4if.org
Subject: Re: [M4IF Technotes] Is standard conform to put mp3 sound in a
MPEG-4 stream ?
Hi,
The objectTypeIndication for mp3 audio is 0x6B as per Table 8 of ISO/IEC
14496-1:2001(E).
Hope this helps u.
rgds,
Ruturaj Anand Chandpur
----- Original Message -----
From: Remigijus
To: technotes@lists.m4if.org
Sent: Wednesday, August 21, 2023 2:58 PM
Subject: [M4IF Technotes] Is standard conform to put mp3 sound in a MPEG-4
stream ?
Hi All
I would like to ask: Is standard conform to put mp3 sound in a MPEG-4
stream:
I've created mp4 file with mp4creator (MPEG4IP) with mp3 sound track inside.
When I'm playing this file with QuickTime 6 It generates Error message:
"File Contains invalid audio data:
An audio Decoder Configuration descriptor corrupted.
Expected ObjectType = Audio(0x40); found object type ^0 instead."
Does it means that QuickTime doesn't supports mp3 sound inside mp4 or
something else.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020821/b342d940/attachment.html
From singer apple.com Wed Aug 21 10:46:39 2002
From: singer apple.com (Dave Singer)
Date: Wed Jul 30 14:08:37 2003
Subject: [M4IF Technotes] Is standard conform to put mp3 sound in a
MPEG-4 stream ?
In-Reply-To: <001201c248f5$0d693020$3301a8c0@technopark.dom>
References: <001201c248f5$0d693020$3301a8c0@technopark.dom>
Message-ID:
At 11:28 +0200 8/21/02, Remigijus wrote:
>Hi All
>
>I would like to ask: Is standard conform to put mp3 sound in a MPEG-4 stream:
>
>I've created mp4 file with mp4creator (MPEG4IP) with mp3 sound track
>inside. When I'm playing this file with QuickTime 6 It generates
>Error message:
>
>
>
>"File Contains invalid audio data:
>
>An audio Decoder Configuration descriptor corrupted.
>
>Expected ObjectType = Audio(0x40); found object type ^0 instead."
>
>
>
>Does it means that QuickTime doesn't supports mp3 sound inside
>mp4 or something else.
>
>
Well yes, it does mean that QT6 doesn't do layer 3 in MP4. QT6
supports a few profiles in MP4, and we currently don't cover the
'old' formats (MPEG1 and MPEG2) in an MP4 context.
I'd be curious to know, however, if we have the same understanding of
the structures which would say "this is layer3 audio". can you
briefly summarize your understanding of the profile, level,
streamtype, objecttype, decoderspecificinfo, and access unit format?
--
David Singer
Apple Computer/QuickTime
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020821/8edda840/attachment.html
From peter mltc.com.tw Thu Aug 22 10:47:53 2002
From: peter mltc.com.tw (Peter Hsu)
Date: Wed Jul 30 14:08:37 2003
Subject: [M4IF Technotes] About short video header.....
Message-ID: <005701c2497d$ee2a89a0$5001a8c0@mltc.com.tw>
Hi All,
The ISO reference code can't encode short video header bitstream.
Where may I get short video header encoder and bitstream?
Thanks.
Peter Hsu
From rchandpur tataelxsi.co.in Thu Aug 22 16:03:35 2002
From: rchandpur tataelxsi.co.in (RUTURAJ CHANDPUR )
Date: Wed Jul 30 14:08:37 2003
Subject: [M4IF Technotes] MP4 hint track
Message-ID: <007401c249be$fd174610$0314010a@tataelxsi.co.in>
Hi,
Can I create MP4 file containing only hint tracks(rtp hint), OD and
BIFS track? If yes, what are the hint track settings?
How can I test this MP4 file?
Thanks in advance.
Ruturaj Anand Chandpur
Sr. Engineer,
Tata Elxsi Ltd.,
www.tataelxsi.com
From manu.batura patni.com Fri Aug 23 16:46:28 2002
From: manu.batura patni.com (Manu V Batura)
Date: Wed Jul 30 14:08:37 2003
Subject: [M4IF Technotes] MPEG-4 bitstreams for conformance tests
Message-ID: <000201c24a8e$259c31c0$5402a8c0@patni.com>
Hi,
From gajananl yahoo.com Sat Aug 24 04:10:29 2002
From: gajananl yahoo.com (gajanan lonkar)
Date: Wed Jul 30 14:08:37 2003
Subject: [M4IF Technotes] Doubts!!
Message-ID: <20020824101029.6902.qmail@web13108.mail.yahoo.com>
Respected Sir(s),
My name is Gajanan Lonkar & i m doing my Final Year
Engg in E&TC from C.O.E.P(INDIA) which is one of the
most reputed colleges in the country. I am doing a
Final Year project at CIRRUS LOGIC titled "MPEG4
SYSTEM PARSER".Before jumping to MPEG4 we were asked
to go through the MPEG1 & MPEG2 ISO Standards.We have
been asked to find the answer to 2 questions but after
much deliberation (and searching on the net) we were
not able to come up with a substantial solution . I
would be grateful if u could throw some light on the
matter.
The questions are as follows
1) Why DTS is Requires
Please note : in our discussion with our guide we
gave a few reasons as to DTS for decoding instant and
PTS for presentation instant but their argument was
that we know the (I,P,B) picture order so we can
decode them just by reordering algorithm. Is it
related to BUFFER MANAGMENT?
2)What is VBV(Video Buffer Verifier)
Where is it located, What parmeters govern its
operation.
Specifically the lower and higher threshold.
PS-We think its operation is somehow related to the
DTS concept.
I hope that u will spare some of ur valuable time and
enlighten us regarding the above queries.
THANKING U IN ANTICIPATION
Your's Sincerely
Gajanan Lonkar
PS- PLease note that the quesions are generic in
manner and can be related to any MPEG version.
__________________________________________________
Do You Yahoo!?
Yahoo! Finance - Get real-time stock quotes
http://finance.yahoo.com
From jaganan hotmail.com Sat Aug 24 16:25:50 2002
From: jaganan hotmail.com (Gajanan)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] Doubts!!
Message-ID:
Respected Sir(s),
My name is Gajanan Lonkar & i m doing my Final Year Engg in E&TC from C.O.E.P(INDIA) which is one of the most reputed colleges in the country. I am doing a Final Year project at CIRRUS LOGIC titled "MPEG4 SYSTEM PARSER".Before jumping to MPEG4 we were asked to go through the MPEG1 & MPEG2 ISO Standards.We have been asked to find the answer to 2 questions but after much deliberation (and searching on the net) we were not able to come up with a substantial solution . I would be highly obliged if u could throw some light on the matter.
The questions are as follows:
1) Why DTS is Requires
Please note : in our discussion with our guide we gave a few reasons as to DTS for decoding instant and PTS for presentation instant but their argument was that we know the (I,P,B)picture order so we can decode them just by reordering. Is it related to BUFFER MANAGMENT?
2)What is VBV(Video Buffer Verifier)
Where is it located, What parmeters govern its operation.
Specifically the lower and higher threshold.
PS-We think its operation is somehow related to the DTS concept.
I hope that u will spare some of ur valuable time and enlighten us regarding the above queries.
THANKING U IN ANTICIPATION
Your's Sincerely
Gajanan Lonkar
PS- PLease note that the quesions are generic in manner and can be related to any MPEG version.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020824/a52ec9f3/attachment.html
From hanxy magima.com.cn Mon Aug 26 11:50:54 2002
From: hanxy magima.com.cn (hanxy)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic model in detail?
Message-ID: <000e01c24cab$65a18d00$b800000a@sh.magima.com>
Hi, All, the description of "bval" in psychoacoustic model of MPEG-4 AAC is "the median bark value of the specific partion", I don't know the physical meaning of it. So, Can you do me a favor to explain it in detail?
Thanks in advance.
your secerely
Speer Han
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020826/508dddb0/attachment.html
From st018487 ait.ac.th Mon Aug 26 12:10:04 2002
From: st018487 ait.ac.th (Suthinee Thanapirom)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] (no subject)
Message-ID: <006d01c24cb6$74d65e80$47129fcb@tc04>
In VTC mode, what is the quantization step size or quantization matrix?
How does arithmetic coding do?Where do it get the probability table from?
Can anyone do me a favor by answering these questions.
Thank you in advance for your help.
Regards,
Suthinee
Email: st018487@ait.ac.th or suthinee_t@hotmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020826/51d46a0c/attachment.html
From idimkovic nero.com Mon Aug 26 10:59:39 2002
From: idimkovic nero.com (Ivan Dimkovic)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic model in detail?
In-Reply-To: <000e01c24cab$65a18d00$b800000a@sh.magima.com>
Message-ID:
Hi,
"BVAL" value represents mean value in bark scale of the specified
partition - i.e. this is the frequency in bark units of the middle of the
partition. For bark units, you need to consult some psychoacoustic books - I
suggest to take a look at Zwicker's publications.
Bark scale is non-linear frequency scale (and Hz is linear frequency scale)
and Bark units correlate to bandpass frequencies of human hearing - there
are several approximation formulas for conversion of Hertz (Hz) to Bark.
Take a look at google for these.
Best regards / Mit freundlichen Gr��?en
Ivan Dimkovic
----------------------------------------------------------------
Ivan Dimkovic
MPEG-4 Software R&D
Ahead Software AG phone: +49 (0)7248 911 822(direct line)
Im Stoeckmaedle 18 fax: +49 (0)7248 911 888
76307 Karlsbad email: idimkovic@nero.com
Germany web: www.nero.com
----------------------------------------------------------------
-----Original Message-----
From: technotes-admin@lists.m4if.org
[mailto:technotes-admin@lists.m4if.org]On Behalf Of hanxy
Sent: Monday, August 26, 2023 4:51 AM
To: technotes@lists.m4if.org
Subject: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic model
in detail?
Hi, All, the description of "bval" in psychoacoustic model of MPEG-4 AAC is
"the median bark value of the specific partion", I don't know the physical
meaning of it. So, Can you do me a favor to explain it in detail?
Thanks in advance.
your secerely
Speer Han
From hanxy magima.com.cn Mon Aug 26 18:46:35 2002
From: hanxy magima.com.cn (hanxy)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic model in detail?
References:
Message-ID: <001e01c24ce5$778b8810$b800000a@sh.magima.com>
----- Original Message -----
From: "Ivan Dimkovic"
To: "hanxy" ;
Sent: Monday, August 26, 2023 3:59 PM
Subject: RE: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic
model in detail?
> Hi,
>
> "BVAL" value represents mean value in bark scale of the specified
> partition - i.e. this is the frequency in bark units of the middle of the
> partition. For bark units, you need to consult some psychoacoustic books -
I
> suggest to take a look at Zwicker's publications.
>
> Bark scale is non-linear frequency scale (and Hz is linear frequency
scale)
> and Bark units correlate to bandpass frequencies of human hearing - there
> are several approximation formulas for conversion of Hertz (Hz) to Bark.
> Take a look at google for these.
>
>
> Best regards / Mit freundlichen Gr��?en
> Ivan Dimkovic
> ----------------------------------------------------------------
> Ivan Dimkovic
> MPEG-4 Software R&D
> Ahead Software AG phone: +49 (0)7248 911 822(direct line)
> Im Stoeckmaedle 18 fax: +49 (0)7248 911 888
> 76307 Karlsbad email: idimkovic@nero.com
> Germany web: www.nero.com
> ----------------------------------------------------------------
>
>
> -----Original Message-----
> From: technotes-admin@lists.m4if.org
> [mailto:technotes-admin@lists.m4if.org]On Behalf Of hanxy
> Sent: Monday, August 26, 2023 4:51 AM
> To: technotes@lists.m4if.org
> Subject: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic
model
> in detail?
>
>
> Hi, All, the description of "bval" in psychoacoustic model of MPEG-4 AAC
is
> "the median bark value of the specific partion", I don't know the physical
> meaning of it. So, Can you do me a favor to explain it in detail?
>
> Thanks in advance.
>
> your secerely
> Speer Han
>
>
From hanxy magima.com.cn Mon Aug 26 18:47:33 2002
From: hanxy magima.com.cn (hanxy)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic model in detail?
References:
Message-ID: <002301c24ce5$9a44ff30$b800000a@sh.magima.com>
Thank a lot, I have calculated and get the median bark value of each
partition!
----- Original Message -----
From: "Ivan Dimkovic"
To: "hanxy" ;
Sent: Monday, August 26, 2023 3:59 PM
Subject: RE: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic
model in detail?
> Hi,
>
> "BVAL" value represents mean value in bark scale of the specified
> partition - i.e. this is the frequency in bark units of the middle of the
> partition. For bark units, you need to consult some psychoacoustic books -
I
> suggest to take a look at Zwicker's publications.
>
> Bark scale is non-linear frequency scale (and Hz is linear frequency
scale)
> and Bark units correlate to bandpass frequencies of human hearing - there
> are several approximation formulas for conversion of Hertz (Hz) to Bark.
> Take a look at google for these.
>
>
> Best regards / Mit freundlichen Gr��?en
> Ivan Dimkovic
> ----------------------------------------------------------------
> Ivan Dimkovic
> MPEG-4 Software R&D
> Ahead Software AG phone: +49 (0)7248 911 822(direct line)
> Im Stoeckmaedle 18 fax: +49 (0)7248 911 888
> 76307 Karlsbad email: idimkovic@nero.com
> Germany web: www.nero.com
> ----------------------------------------------------------------
>
>
> -----Original Message-----
> From: technotes-admin@lists.m4if.org
> [mailto:technotes-admin@lists.m4if.org]On Behalf Of hanxy
> Sent: Monday, August 26, 2023 4:51 AM
> To: technotes@lists.m4if.org
> Subject: [M4IF Technotes] Can anyone explain "bval" in psychoacoustic
model
> in detail?
>
>
> Hi, All, the description of "bval" in psychoacoustic model of MPEG-4 AAC
is
> "the median bark value of the specific partion", I don't know the physical
> meaning of it. So, Can you do me a favor to explain it in detail?
>
> Thanks in advance.
>
> your secerely
> Speer Han
>
>
From sunx pollux.usc.edu Mon Aug 26 01:30:15 2002
From: sunx pollux.usc.edu (sunx)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] FGS parameter file example
Message-ID:
Dear MPEG-4 users,
Can anybody send me one copy of the parameter file used for MPEG-4 FGS
coding? I am confused by those parameters defined in the parameter file.
Thanks,
Xiaoming
From sunx pollux.usc.edu Mon Aug 26 12:45:43 2002
From: sunx pollux.usc.edu (sunx)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] Subscribe to mpeg.nist.gov
Message-ID:
Dear MPEG-4 experts,
My group subscribed to mpeg.nist.gov before, but later my advisor decided
to cancel the subscription. Can anyone tell me how to subscribe to
mpeg.nist.gov and how much is the cost for one year subscription?
Thanks,
Xiaoming
From rkoenen intertrust.com Mon Aug 26 12:51:51 2002
From: rkoenen intertrust.com (Rob Koenen)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] Subscribe to mpeg.nist.gov
Message-ID: <3C124172E7FDD511B510000347426D59011B38EA@exchange.epr.com>
As I already mentioned in a personal mail, M4IF does not deal with
those issues. Please contact Peter Schirling, the head of the US
MPEG delegation, as originally suggessted.
Kind Regards,
Rob
> -----Original Message-----
> From: sunx [mailto:sunx@pollux.usc.edu]
> Sent: Monday, August 26, 2023 11:46
> To: MPEG-4 Technotes
> Subject: [M4IF Technotes] Subscribe to mpeg.nist.gov
>
>
>
> Dear MPEG-4 experts,
>
> My group subscribed to mpeg.nist.gov before, but later my
> advisor decided
> to cancel the subscription. Can anyone tell me how to subscribe to
> mpeg.nist.gov and how much is the cost for one year subscription?
>
> Thanks,
>
> Xiaoming
>
>
>
> _______________________________________________
> Technotes mailing list
> Technotes@lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>
From jwalant.desai wipro.com Tue Aug 27 16:17:37 2002
From: jwalant.desai wipro.com (Jwalant Sumantrai Desai)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] H.264 Conformance testing
Message-ID:
Skipped content of type multipart/alternative-------------- next part --------------
**************************Disclaimer************************************
Information contained in this E-MAIL being proprietary to Wipro Limited is
'privileged' and 'confidential' and intended for use only by the individual
or entity to which it is addressed. You are notified that any use, copying
or dissemination of the information contained in the E-MAIL in any manner
whatsoever is strictly prohibited.
***************************************************************************
From nakata mpeg.rcast.u-tokyo.ac.jp Tue Aug 27 23:24:17 2002
From: nakata mpeg.rcast.u-tokyo.ac.jp (Mayumi (Nakata) Koike)
Date: Wed Jul 30 14:08:38 2003
Subject: [M4IF Technotes] MPEG-4 bitstreams for conformance tests
In-Reply-To: <000201c24a8e$259c31c0$5402a8c0@patni.com>
Message-ID: <4.3.2-J.20020827221918.05ee6960@po.mpeg.rcast.u-tokyo.ac.jp>
Dear Manu,
Hi, here's the answer for your question.
Host Name po.mpeg.rcast.u-tokyo.ac.jp
User ID mpegbs
Password Mp4-1999
Then, enter "public_htm" folder to find bitstreams.
Please use FTP software to download.
Best,
Mayumi KOIKE
At 15:46 02/08/23 +0530, Manu V Batura wrote:
>Hi,
>
> From where can one procure the bitstreams donated by MPEG-4 Platform
>Verification Bitstream Development Project of Japan?
>Are they in public domain?
>
>Thanks & Regards,
>Manu
>
>
>_______________________________________________
>Technotes mailing list
>Technotes@lists.m4if.org
>http://lists.m4if.org/mailman/listinfo/technotes
From rkoenen intertrust.com Wed Aug 28 13:32:55 2002
From: rkoenen intertrust.com (Rob Koenen)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] H.264 Conformance testing
Message-ID: <3C124172E7FDD511B510000347426D59011B396B@exchange.epr.com>
Jwalant,
The H.264 / MPEG-4 AVC standard itself isn't finalized yet.
Conformance always comes later than the standard,
usually by 6-9 months.
Rob
-----Original Message-----
From: Jwalant Sumantrai Desai [mailto:jwalant.desai@wipro.com]
Sent: Tuesday, August 27, 2023 2:48
To: technotes@lists.m4if.org
Subject: [M4IF Technotes] H.264 Conformance testing
Hi all,
I wish to know if the conformance testing criteria for H.264 has been
finalized?
If Yes, it would be helpful if someone could provide some pointers to
availability of the compliance testing standard and test bitstreams for
H.264.
If No, what is the likely date for availability of the same.
Thanks and rgds
Jwalant
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20020828/7daa5201/attachment.html
From praveen smartyantra.com Thu Aug 29 10:48:27 2002
From: praveen smartyantra.com (Praveen Patil)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] Testing H.263 Encoder
Message-ID: <3D6DA093.000006.01080@sindhu>
Skipped content of type multipart/alternative-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 494 bytes
Desc: not available
Url : /pipermail/mp4-tech/attachments/20020829/68afe9dd/attachment.gif
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/jpeg
Size: 1431 bytes
Desc: not available
Url : /pipermail/mp4-tech/attachments/20020829/68afe9dd/attachment.jpeg
From byan cse.cuhk.edu.hk Thu Aug 29 21:32:55 2002
From: byan cse.cuhk.edu.hk (Yan Bo)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] about the NTT DoCoMo software
Message-ID: <200208291232.g7TCWru21351@cse.cuhk.edu.hk>
Dear all,
Can anyone tell me where to download the NTT DoCoMo software, which can insert many kinds of errors in the compressed video bitstream?
Thanks and Regards!
Yan Bo
From Jeyaprakash.S lntinfotech.com Thu Aug 29 18:22:26 2002
From: Jeyaprakash.S lntinfotech.com (Jeyaprakash.S@lntinfotech.com)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] CELP Encoder
Message-ID:
Hi,
I am having a doubt in ISO/IEC 14496-3:2001(E) Annex 3.B MPEG-4 CELP
encoder tools.
the n_lpc_analysis value for 8 KHz is given as (sbfrm_size/80)
Can you please confirm whether it is sbfrm_size or frame_size.
Warm Regards,
Jeyaprakash S
Jeyaprakash.S
Software Engineer( Communications and Embedded Systems )L&T Infotech Ltd.
Prestige Commercial Complex
4th Floor
#2,Church Street
Bangalore - 560 001, INDIA
Off : +91 080 532 3734 /3736 Extn : 350
From mylanchen yahoo.com Thu Aug 29 06:11:43 2002
From: mylanchen yahoo.com (xin chen)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] questions about mpeg-4 file format
Message-ID: <20020829121143.47319.qmail@web12101.mail.yahoo.com>
Hi All,
I read the related part in ISO/IEC 14496-1, and
quicktime file format, but they are all concentrate on
the meta-data part, that is the structure of moov
atom. I want to know more about how the samples are
stored (in mdat atom, right). Where to get documents?
And about the SL layer, since the SL packet hearder is
composed according to the SLconfigdescriptor, which is
stored in ESDescriptor, my question is: the SL packet
hearder are generated only when the mpeg4 file is
transmitted, or is it stored already in the .mp4
(Because on the figure 1 of ISO/IEC 14496-1, the
terminal get a file from storage medium will need to
go through deliver layer and sync layer for decoding
too.)? I think it's a waste to store the same inf.
twice.
Thank you very much.
__________________________________________________
Do You Yahoo!?
Yahoo! Finance - Get real-time stock quotes
http://finance.yahoo.com
From mjacklin geneva-link.ch Thu Aug 29 17:40:42 2002
From: mjacklin geneva-link.ch (Martin Jacklin)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] Doing any demos at IBC?
Message-ID:
Hi
This message is for anyone, member or non-member, who is showing MPEG-4 applications at IBC in a couple of weeks, who hasn't yet provided me with their IBC demo description,...
(This also applies if you are presenting a paper)
Could you send it to me for some free publicity in our MPEG-4 at IBC exhibit guide?
I can fit 30 words. I need it yesterday, but am holding open the gate specially for you, until tomorrow.
/\/\artin Jacklin
Hypermedium
(M4IF marketing support and media relations)
martin.jacklin@m4if.org
+41 79 291 1882
gsm/sms/fax
From richard sorenson.com Thu Aug 29 10:10:28 2002
From: richard sorenson.com (Richard Shields)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] questions about mpeg-4 file format
In-Reply-To: <20020829121143.47319.qmail@web12101.mail.yahoo.com>
Message-ID: <73FC8F68-BB61-11D6-8C4A-0030657E211C@sorenson.com>
Hi xin chen,
The mdat atom can contain any data in any order you see fit. There is
no "framing" data stored in the mdat and thus it has no structure. It
is the meta data in the moov atom that gives the data in the mdat atom
structure but only indirectly. Typically you would write video or
audio samples consecutively in decode order into the mdat but there are
no rules stating that you have too. You have both audio and video
tracks it's advantageous to write them in an interleaved fashion. The
moov atom will indicate to the player how to locate the samples.
SL Layer... I'll let the more educated among us to answer this one :-)
-Richard
On Thursday, August 29, 2002, at 06:11 AM, xin chen wrote:
> Hi All,
> I read the related part in ISO/IEC 14496-1, and
> quicktime file format, but they are all concentrate on
> the meta-data part, that is the structure of moov
> atom. I want to know more about how the samples are
> stored (in mdat atom, right). Where to get documents?
>
> And about the SL layer, since the SL packet hearder is
> composed according to the SLconfigdescriptor, which is
> stored in ESDescriptor, my question is: the SL packet
> hearder are generated only when the mpeg4 file is
> transmitted, or is it stored already in the .mp4
> (Because on the figure 1 of ISO/IEC 14496-1, the
> terminal get a file from storage medium will need to
> go through deliver layer and sync layer for decoding
> too.)? I think it's a waste to store the same inf.
> twice.
>
> Thank you very much.
>
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Finance - Get real-time stock quotes
> http://finance.yahoo.com
> _______________________________________________
> Technotes mailing list
> Technotes@lists.m4if.org
> http://lists.m4if.org/mailman/listinfo/technotes
>
From emre.aksu nokia.com Thu Aug 29 19:04:37 2002
From: emre.aksu nokia.com (emre.aksu@nokia.com)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] Visual Object Sequence Start Code
Message-ID:
Hello,
I'd like to ask a question related to the Visual Object Sequence:
In an MPEG-4 visual bitstream, can there be more than one Visual Object Sequence Start Code coming one after another without having a Visual Object Sequence End Code to terminate the previous one, i.e. :
visual_object_sequence_start_code
...
...
visual_object_sequence_start_code
...
...
visual_object_sequence_end_code
Looking forward to receiving an answer,
Best Regards,
Emre Baris Aksu
Nokia Corp.
From peter mltc.com.tw Fri Aug 30 11:01:21 2002
From: peter mltc.com.tw (Peter Hsu)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] Visual Object Sequence Start Code
References:
Message-ID: <003401c24fc9$2421a240$5001a8c0@mltc.com.tw>
Hi, Emre Baris Aksu,
Yes, you can include more than one Visual Object Sequence Start Code before
a Visual Object Sequence End Code.
Peter Hsu
MicroLinks Technology Corp.
21F, No.55, Chung Cheng 3rd RD., Kaohsiung, Taiwan 800, R.O.C.
Tel: 886-7-2225678 Ext. 534
Fax: 886-7-2225675
E-mail: peter@mltc.com.tw
----- Original Message -----
From:
To:
Sent: Thursday, August 29, 2023 11:04 PM
Subject: [M4IF Technotes] Visual Object Sequence Start Code
Hello,
I'd like to ask a question related to the Visual Object Sequence:
In an MPEG-4 visual bitstream, can there be more than one Visual Object
Sequence Start Code coming one after another without having a Visual Object
Sequence End Code to terminate the previous one, i.e. :
visual_object_sequence_start_code
...
...
visual_object_sequence_start_code
...
...
visual_object_sequence_end_code
Looking forward to receiving an answer,
Best Regards,
Emre Baris Aksu
Nokia Corp.
_______________________________________________
Technotes mailing list
Technotes@lists.m4if.org
http://lists.m4if.org/mailman/listinfo/technotes
From neel2265 yahoo.com Fri Aug 30 00:08:30 2002
From: neel2265 yahoo.com (Neelakanth)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] Access unit reordering
Message-ID: <20020830060830.57811.qmail@web40206.mail.yahoo.com>
Hi All
I am going through the ISMA Implementation
Specification Version 1.0 (dated:28 August 2001).
In appendix A of the document it is explained that "To
reconstruct the original order, the RTP
time stamp and the AU-Index-delta are used." This is
possible if all the Access Units (AU's)
are of same time duration. I would be obliged to you,
if you let me know the strategy for
reordering the AU's on the Client side if the Access
Units (AU's) are of variable time duration.
Thanks
Neelakanth
__________________________________________________
Do You Yahoo!?
Yahoo! Finance - Get real-time stock quotes
http://finance.yahoo.com
From singer apple.com Fri Aug 30 17:32:08 2002
From: singer apple.com (Dave Singer)
Date: Wed Jul 30 14:08:39 2003
Subject: [M4IF Technotes] questions about mpeg-4 file format
In-Reply-To: <73FC8F68-BB61-11D6-8C4A-0030657E211C@sorenson.com>
References: <73FC8F68-BB61-11D6-8C4A-0030657E211C@sorenson.com>
Message-ID:
Richard has it perfect for the media data. The access units are
stored in the mdat in the placesyou compute from the chunk offsets
and sample sizes, and the sample sizes give you, um, the sizes.
The SL headers are not stored in the file. There is a pre-defined
value of the SLConfig (2) that is used in the file format. It is
expected that any hinter/streamer that delivers the data using SL
headers will replace this with an appropriate config that matches
what they do. The pre-defined value means essentially 'you're in a
file, the info that would normally be in SL headers is in the
meta-data'.
At 09:10 -0600 8/29/02, Richard Shields wrote:
>Hi xin chen,
>
>The mdat atom can contain any data in any order you see fit. There
>is no "framing" data stored in the mdat and thus it has no
>structure. It is the meta data in the moov atom that gives the data
>in the mdat atom structure but only indirectly. Typically you would
>write video or audio samples consecutively in decode order into the
>mdat but there are no rules stating that you have too. You have
>both audio and video tracks it's advantageous to write them in an
>interleaved fashion. The moov atom will indicate to the player how
>to locate the samples.
>
>SL Layer... I'll let the more educated among us to answer this one :-)
>
>-Richard
>
>On Thursday, August 29, 2002, at 06:11 AM, xin chen wrote:
>
>>Hi All,
>>I read the related part in ISO/IEC 14496-1, and
>>quicktime file format, but they are all concentrate on
>>the meta-data part, that is the structure of moov
>>atom. I want to know more about how the samples are
>>stored (in mdat atom, right). Where to get documents?
>>
>>And about the SL layer, since the SL packet hearder is
>>composed according to the SLconfigdescriptor, which is
>>stored in ESDescriptor, my question is: the SL packet
>>hearder are generated only when the mpeg4 file is
>>transmitted, or is it stored already in the .mp4
>>(Because on the figure 1 of ISO/IEC 14496-1, the
>>terminal get a file from storage medium will need to
>>go through deliver layer and sync layer for decoding
>>too.)? I think it's a waste to store the same inf.
>>twice.
>>
>>Thank you very much.
>>
>>__________________________________________________
>>Do You Yahoo!?
>>Yahoo! Finance - Get real-time stock quotes
>>http://finance.yahoo.com
>>_______________________________________________
>>Technotes mailing list
>>Technotes@lists.m4if.org
>>http://lists.m4if.org/mailman/listinfo/technotes
>>
>_______________________________________________
>Technotes mailing list
>Technotes@lists.m4if.org
>http://lists.m4if.org/mailman/listinfo/technotes
--
David Singer
Apple Computer/QuickTime