From s2318008 ntut.edu.tw Mon May 2 07:07:02 2005 From: s2318008 ntut.edu.tw (=?big5?B?rNPZeShIdWFuZy1CaW4p?=) Date: Mon May 2 08:40:53 2005 Subject: [Mp4-tech] about short-term/long-term PicNum question Message-ID: <013101c54e9a$1a7c21c0$04000100@chan> Dear experts, The frame_num is incremented by 1 in modulo MaxFrameNum operation. There is a example in standard as follows. "For example, when three reference frames are marked as "used for short-term reference" with PicNum equal to 300, 302, and 303 and two reference frames are marked as "used for long-term reference" with LongTermPicNum equal to 0 and 3, the initial index order is: - RefPicList0[0] is set equal to the short-term reference picture with PicNum = 303, - RefPicList0[1] is set equal to the short-term reference picture with PicNum = 302, - RefPicList0[2] is set equal to the short-term reference picture with PicNum = 300, - RefPicList0[3] is set equal to the long-term reference picture with LongTermPicNum = 0, and - RefPicList0[4] is set equal to the long-term reference picture with LongTermPicNum = 3." In this example, we have different PicNum whatever the reference frame is a shot-term or long-term picuture. However, if the example is modified as follows. "For example, when three reference frames are marked as "used for short-term reference" with PicNum equal to 4, 3, 2and and one reference frame is marked as "used for long-term reference" with LongTermPicNum equal to 2, the initial index order is: - RefPicList0[0] is set equal to the short-term reference picture with PicNum = 4, - RefPicList0[1] is set equal to the short-term reference picture with PicNum = 3, - RefPicList0[2] is set equal to the short-term reference picture with PicNum = 2, - RefPicList0[3] is set equal to the long-term reference picture with LongTermPicNum = 2" Is this example possible? If yes, how should we pick up the wanted frame to be the reference frame when there are two picture owning the same PicNum in both short-term and long-term memory ? Thx H.B. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050502/4683cc79/attachment.html From janisjp hanmail.net Mon May 2 19:36:17 2005 From: janisjp hanmail.net (janis) Date: Mon May 2 08:40:58 2005 Subject: [Mp4-tech] Distortion Prediction in JVT-O016 Message-ID: <20050502183617.HM.00000000000MGAF@wwl27.hanmail.net> An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050502/5cda255c/attachment.html From bharatsoni gmail.com Mon May 2 17:58:19 2005 From: bharatsoni gmail.com (Bharat Soni) Date: Mon May 2 08:41:04 2005 Subject: [Mp4-tech] SAD-PSNR (currrent/reference-original/reconstructed) In-Reply-To: <20050427144344.HM.00000000000MEFV@wwl27.hanmail.net> References: <20050427144344.HM.00000000000MEFV@wwl27.hanmail.net> Message-ID: Hi Janis, I think there is a little bit confusion in which frame(original/reconstructed) is used when. Lets consider following frames:I"1: Original Frame to be coded as I frameI1: is reconstructed (decoded) frame that will be used as reference frame.P"2: original frame which has to be predicted from I1 and coded as P frame. You may (need to) determine the encoding quality of I"1 only when ithas already been compressed and again decompressed to generatereconstructed frame. So to determine the quality of encoding (PSNR),contents (block data) of reconstructed I1 will be compared with theoriginal data of I1 i.e. I"1. (I"1 <=> I1) For encoding P1, we use the RECONSTRUCTED I1 and NOT the original I"1frame as at the decoder side, we do not have the original frameavailable for motion compensation.So on the encoder side for forming prediction, we will compare thecontents of the current (original) frame data with the contents of thereference (reconstructed) frame to determine SAD/MAD etc..(P"2 <=> I1). Regarding relation between SAD and PSNR, I would say there is inverserelation between the two. lower SAD means the blocks are verysimilar, this in turn means lower error values of the coefficient thatmeans lower the chances of the error due to quantization (thisstatement may not be true always, but this is how it works). Smallererror means the reconstructed frame will be close to original hence itgives high PSNR value, which inturn represent good encoding quality. I hope its clear now. Regards,Bharat On 4/27/05, janis wrote:> Dear all, > > In the previous question about SAD and PSNR, > what I really want to know is this, > > To get a SAD, we compare the currnet frame block with a reference frame> block. > But to get MSE or PSNR as a measurement for the quality, > we compare the reconstructed frame with the original frame. > > But there ARE some relation between SAD and PSNR, > that when we find the smallest SAD, we can get better quality. > > So, why is this happening? > How can I tell the relation between comparing 'the original with the> reconstructed' > and 'the current with reference'? > > Hope I made my question clear, > Thanks in advance, > janis > > > "?? ???, Daum" http://www.daum.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 rfarley qualcomm.com Mon May 2 09:59:00 2005 From: rfarley qualcomm.com (Richard Farley) Date: Mon May 2 12:22:23 2005 Subject: [Mp4-tech] MPEG2 TS to YUV/AVI, PSNR tools Message-ID: <6.2.2.1.2.20050502085649.11a80280@unixmail.qualcomm.com> Hi, I'm looking for tools to do MPEG2 transport stream to YUV/AVI conversion. I'm also looking for tools to do PSNR and other video quality measurements. Thanks in advance, Richard From garysull windows.microsoft.com Mon May 2 17:21:17 2005 From: garysull windows.microsoft.com (Gary Sullivan) Date: Tue May 3 04:52:03 2005 Subject: [Mp4-tech] about short-term/long-term PicNum question Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2460E55B09E@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> Yes, such an example could be possible. Regarding your second question, in this example there is only one picture with a particular value of PicNum. There is a picture with a value of LongTermPicNum that is equal to the value of PicNum on another picture, but this is irrelevant. PicNum and LongTermPicNum are two different variables. You should probably study more closely the derivation of the variable LongTermPicNum. For example, are you aware that the value of frame_num is irrelevant to the value of LongTermPicNum? Best regards, Gary Sullivan ________________________________ From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of ??(Huang-Bin) Sent: Sunday, May 01, 2024 3:07 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] about short-term/long-term PicNum question Dear experts, The frame_num is incremented by 1 in modulo MaxFrameNum operation. There is a example in standard as follows. "For example, when three reference frames are marked as "used for short-term reference" with PicNum equal to 300, 302, and 303 and two reference frames are marked as "used for long-term reference" with LongTermPicNum equal to 0 and 3, the initial index order is: - RefPicList0[0] is set equal to the short-term reference picture with PicNum = 303, - RefPicList0[1] is set equal to the short-term reference picture with PicNum = 302, - RefPicList0[2] is set equal to the short-term reference picture with PicNum = 300, - RefPicList0[3] is set equal to the long-term reference picture with LongTermPicNum = 0, and - RefPicList0[4] is set equal to the long-term reference picture with LongTermPicNum = 3." In this example, we have different PicNum whatever the reference frame is a shot-term or long-term picuture. However, if the example is modified as follows. "For example, when three reference frames are marked as "used for short-term reference" with PicNum equal to 4, 3, 2and and one reference frame is marked as "used for long-term reference" with LongTermPicNum equal to 2, the initial index order is: - RefPicList0[0] is set equal to the short-term reference picture with PicNum = 4, - RefPicList0[1] is set equal to the short-term reference picture with PicNum = 3, - RefPicList0[2] is set equal to the short-term reference picture with PicNum = 2, - RefPicList0[3] is set equal to the long-term reference picture with LongTermPicNum = 2" Is this example possible? If yes, how should we pick up the wanted frame to be the reference frame when there are two picture owning the same PicNum in both short-term and long-term memory ? Thx H.B. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050502/104218a8/attachment.html From Galina.Sabeva Micronas.com Tue May 3 12:41:55 2005 From: Galina.Sabeva Micronas.com (Sabeva Galina) Date: Wed May 4 07:21:59 2005 Subject: [Mp4-tech] create artifical test sequence Message-ID: Hi all, can someone tell me how can i create artificial test sequence using JM and where i can read about this issue. Thank you in advance. galina -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050503/03e94a2a/attachment.html From pankaj0iit yahoo.co.in Tue May 3 23:11:39 2005 From: pankaj0iit yahoo.co.in (pankaj bansal) Date: Wed May 4 07:25:01 2005 Subject: [Mp4-tech] interlace field support in H.264 Message-ID: <20050504051139.71295.qmail@web8407.mail.in.yahoo.com> I have one doubt related to the decoded reference buffer in the case of interlaced support. Is there only frames are stored or field can be stored? for example suppose ref_buffer[8] is a buffer used to store the decoded ref. frames or fields. Now which of the following thing can be possible (1) ref_buffer[0] store frame1 ref_buffer[1] store top_field(frame2) ref_buffer[2] store bottom_feld(frame2) ........... (2) ref_buffer[0] store frame(frame1) ref_buffer[1] store top_field and bottom_field of (frame2) thanks & regards pankaj __________________________________________________ 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/20050503/cf628eb4/attachment.html From abdessamad.falhi epfl.ch Wed May 4 15:56:55 2005 From: abdessamad.falhi epfl.ch (abdessamad.falhi@epfl.ch) Date: Wed May 4 09:33:46 2005 Subject: [Mp4-tech] documention on aac decoder Message-ID: <1115211415.4278c697a8ca8@imapwww.epfl.ch> Hi, Does anybody know where I can find explanation about mpeg4 aac decoder implementation? I am beginner in this field. regards From abdessamad.falhi epfl.ch Wed May 4 17:48:39 2005 From: abdessamad.falhi epfl.ch (abdessamad.falhi@epfl.ch) Date: Wed May 4 11:25:31 2005 Subject: [Mp4-tech] documention on audio aac decoder In-Reply-To: <1115211415.4278c697a8ca8@imapwww.epfl.ch> References: <1115211415.4278c697a8ca8@imapwww.epfl.ch> Message-ID: <1115218119.4278e0c75d4fc@imapwww.epfl.ch> Hi, Does anybody know where I can find explanation about audio mpeg4 aac decoder implementation? I am beginner in this field. regards From garysull windows.microsoft.com Wed May 4 12:49:11 2005 From: garysull windows.microsoft.com (Gary Sullivan) Date: Wed May 4 17:17:20 2005 Subject: [Mp4-tech] interlace field support in H.264 Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2460E66AB00@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> Fields can be stored. Certain pairs of fields are grouped together in two special ways. Those special pairs are called complementary reference field pairs and complementary non-reference field pairs. You should read the details of the standard to understand how this stuff really works. The basic notion is that the decoder reference picture buffer is considered to be constructed of memory areas that are structured as frames. When appropriate, two fields will be placed in the same decoder frame memory storage area. Fields can only be coupled together for use as a reference frame when this is the case. So the model basically fits your structuring described in "(2)", rather than "(1)". But again I caution you to study and consider all the details before jumping to any overly-general conclusions. Best regards, Gary Sullivan ________________________________ From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of pankaj bansal Sent: Tuesday, May 03, 2024 10:12 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] interlace field support in H.264 I have one doubt related to the decoded reference buffer in the case of interlaced support. Is there only frames are stored or field can be stored? for example suppose ref_buffer[8] is a buffer used to store the decoded ref. frames or fields. Now which of the following thing can be possible (1) ref_buffer[0] store frame1 ref_buffer[1] store top_field(frame2) ref_buffer[2] store bottom_feld(frame2) ........... (2) ref_buffer[0] store frame(frame1) ref_buffer[1] store top_field and bottom_field of (frame2) thanks & regards pankaj __________________________________________________ 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/20050504/45babccc/attachment.html From kaustubh.patankar vsnl.net Thu May 5 09:32:48 2005 From: kaustubh.patankar vsnl.net (kaustubh.patankar@vsnl.net) Date: Thu May 5 08:27:12 2005 Subject: [Mp4-tech] SAD-PSNR (currrent/reference-original/reconstructed) Message-ID: Hi Bharat One correction, there are implementations where original is also used for encoding. with regards Kaustubh ----- Original Message ----- From: Bharat Soni Date: Monday, May 2, 2024 4:58 pm Subject: Re: [Mp4-tech] SAD-PSNR (currrent/reference-original/reconstructed) > Hi Janis, > I think there is a little bit confusion in which > frame(original/reconstructed) is used when. > Lets consider following frames:I"1: Original Frame to be coded as > I frameI1: is reconstructed (decoded) frame that will be used as > reference frame.P"2: original frame which has to be predicted from > I1 and coded as P frame. > You may (need to) determine the encoding quality of I"1 only when > ithas already been compressed and again decompressed to > generatereconstructed frame. So to determine the quality of > encoding (PSNR),contents (block data) of reconstructed I1 will be > compared with theoriginal data of I1 i.e. I"1. (I"1 <=> I1) > For encoding P1, we use the RECONSTRUCTED I1 and NOT the original > I"1frame as at the decoder side, we do not have the original > frameavailable for motion compensation.So on the encoder side for > forming prediction, we will compare thecontents of the current > (original) frame data with the contents of thereference > (reconstructed) frame to determine SAD/MAD etc..(P"2 <=> I1). > Regarding relation between SAD and PSNR, I would say there is > inverserelation between the two. lower SAD means the blocks are > verysimilar, this in turn means lower error values of the > coefficient thatmeans lower the chances of the error due to > quantization (thisstatement may not be true always, but this is > how it works). Smallererror means the reconstructed frame will be > close to original hence itgives high PSNR value, which inturn > represent good encoding quality. > I hope its clear now. > Regards,Bharat > > On 4/27/05, janis wrote:> Dear all, > > In > the previous question about SAD and PSNR, > what I really want to > know is this, > > To get a SAD, we compare the currnet frame block > with a reference frame> block. > But to get MSE or PSNR as a > measurement for the quality, > we compare the reconstructed frame > with the original frame. > > But there ARE some relation between > SAD and PSNR, > that when we find the smallest SAD, we can get > better quality. > > So, why is this happening? > How can I tell > the relation between comparing 'the original with the> > reconstructed' > and 'the current with reference'? > > Hope I made > my question clear, > Thanks in advance, > janis > > > "?? ???, > Daum" http://www.daum.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 hav! > e.> > Note: Conduct on the mailing list is subject to the > Antitrust guidelines> found at> " > target="l">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 From kgshail yahoo.com Thu May 5 03:58:52 2005 From: kgshail yahoo.com (Shailendra Gupta) Date: Thu May 5 08:30:13 2005 Subject: [Mp4-tech] Motion Estimation Message-ID: <20050505095852.44652.qmail@web52608.mail.yahoo.com> Hi; Please can any one suggest the standard ways to test the correctness of Motion Estimation and Motion Compensation vectors just like there are ways for DCT/IDCT, VLCs. Regards Shailendra __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From pankaj0iit yahoo.co.in Thu May 5 04:56:39 2005 From: pankaj0iit yahoo.co.in (pankaj bansal) Date: Thu May 5 08:31:22 2005 Subject: [Mp4-tech] interlace field support in H.264 In-Reply-To: <91D7F2CEE3425A4A9D11311D09FCE2460E66AB00@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> Message-ID: <20050505105640.91631.qmail@web8405.mail.in.yahoo.com> I have one more doubt, In a encoded stream(with config_file setting pic_interlace = 2 and mb_interlace = 0) Is it possible that for P and B slices (excluding mb_type B_Skip, B_Direct_16x16, B_Direct_8x8 ) a MB in a field is predicted from a frame or a MB in a frame is pdredicted from a field? Actually i am confused with the clause 8.4.2.1 in H.264 standard. can you explan this clause more clearly? Thanks & Regards Pankaj Kumar Bansal Gary Sullivan wrote:Fields can be stored. Certain pairs of fields are grouped together in two special ways. Those special pairs are called complementary reference field pairs and complementary non-reference field pairs. You should read the details of the standard to understand how this stuff really works. The basic notion is that the decoder reference picture buffer is considered to be constructed of memory areas that are structured as frames. When appropriate, two fields will be placed in the same decoder frame memory storage area. Fields can only be coupled together for use as a reference frame when this is the case. So the model basically fits your structuring described in "(2)", rather than "(1)". But again I caution you to study and consider all the details before jumping to any overly-general conclusions. Best regards, Gary Sullivan --------------------------------- From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of pankaj bansal Sent: Tuesday, May 03, 2024 10:12 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] interlace field support in H.264 I have one doubt related to the decoded reference buffer in the case of interlaced support. Is there only frames are stored or field can be stored? for example suppose ref_buffer[8] is a buffer used to store the decoded ref. frames or fields. Now which of the following thing can be possible (1) ref_buffer[0] store frame1 ref_buffer[1] store top_field(frame2) ref_buffer[2] store bottom_feld(frame2) ........... (2) ref_buffer[0] store frame(frame1) ref_buffer[1] store top_field and bottom_field of (frame2) thanks & regards pankaj __________________________________________________ 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 --------------------------------- Yahoo! Mail Mobile Take Yahoo! Mail with you! Check email on your mobile phone. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050505/6613e704/attachment.html From poojatiwari16 gmail.com Thu May 5 19:52:56 2005 From: poojatiwari16 gmail.com (Pooja Tiwari) Date: Thu May 5 17:33:59 2005 Subject: [Mp4-tech] Regarding Standard MPEG AAC Decoder Output Message-ID: <28419f130505050622594e9bf7@mail.gmail.com> Hello I have downloaded the ISO Standard Decoder file from the ISO site. However on building the code and running it with standard adif files, it gives a lot of noise. Please Tell me how to deal with the problem Also when ADTS input is given to the Decoder, the output is totally noise and i am unable to understand the reason. Please Help -- ~Pooja From mk_sabeel yahoo.co.in Thu May 5 09:07:59 2005 From: mk_sabeel yahoo.co.in (M.K.Sabeel) Date: Thu May 5 17:34:04 2005 Subject: [Mp4-tech] Mpeg4 - celp encoder Message-ID: <20050505150759.81279.qmail@web8309.mail.in.yahoo.com> Hi Can anyone please let me know where i can find an MPEG4-CELP encoder. i have tried the encoder from mpeg4 audio reference software.wherein i first created celp stream in flexmux form and then converted flexmux to mp4 fileformat,but this mp4 fileformat stream is not played by all the players i have like platform4 player,quicktime,mpegable player etc. thanks sabeel --------------------------------- Do you Yahoo!? Yahoo! Mail - now with 250MB free storage. Learn more. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050505/ec0050df/attachment.html From dsn2603 rediffmail.com Thu May 5 16:32:05 2005 From: dsn2603 rediffmail.com (sakthi narayanan) Date: Thu May 5 17:34:09 2005 Subject: [Mp4-tech] documention on aac decoder Message-ID: <20050505153328.30622.qmail@webmail29.rediffmail.com> Hi abdessamad, Go thru this site http://www.audiocoding.com/modules/wiki/ with regards, sakthi On Wed, 04 May 2024 abdessamad.falhi@epfl.ch wrote : > > > Hi, > > Does anybody know where I can find explanation about mpeg4 aac decoder >implementation? I am beginner in this field. > > regards > >_______________________________________________ >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/20050505/2e9d59bf/attachment.html From w_tong2000 yahoo.com Thu May 5 11:43:02 2005 From: w_tong2000 yahoo.com (Tong Wu) Date: Thu May 5 17:34:14 2005 Subject: [Mp4-tech] mismatch between the reference software and spec on transform_size_8x8_flag Message-ID: <20050505174302.85955.qmail@web50706.mail.yahoo.com> Hi, In the bistream syntax specification the transform_size_8x8_flag syntax is ahead of the coded_block_pattern syntax. However when I read the readCBPandCoeffsFromNAL()in the reference software JM93, I find the coded_block_pattern syntax is parsed before transform_size_8x8_flag. Does anybody notice this mismatch? Thanks. Tony __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From ksuehring web.de Fri May 6 01:48:53 2005 From: ksuehring web.de (Karsten Suehring) Date: Fri May 6 09:27:55 2005 Subject: [Mp4-tech] interlace field support in H.264 In-Reply-To: <20050505105640.91631.qmail@web8405.mail.in.yahoo.com> References: <20050505105640.91631.qmail@web8405.mail.in.yahoo.com> Message-ID: <427AA2D5.7050709@web.de> A frame macroblock in a frame coded picture can be predicted from a reference frame or a complementary reference field pair. A macroblock in a field coded piture can be predicted from single fields, fields of reference field pairs and fields of coded frames. I would suggest reading section 8.2.4.2 "Initialisation process for reference picture lists" Best regards, Karsten pankaj bansal wrote: > > > */I have one more doubt, /*In a encoded stream(with config_file setting > pic_interlace = 2 and mb_interlace = 0) Is it possible that for P and B > slices (excluding mb_type B_Skip, B_Direct_16x16, B_Direct_8x8 ) a MB in > a field is predicted from a frame or a MB in a frame is pdredicted from > a field? > > Actually i am confused with the clause 8.4.2.1 in H.264 standard. > > can you explan this clause more clearly? > > Thanks & Regards > > Pankaj Kumar Bansal > > > > */Gary Sullivan /* wrote: > > Fields can be stored. Certain pairs of fields are grouped together > in two special ways. Those special pairs are called complementary > reference field pairs and complementary non-reference field pairs. > You should read the details of the standard to understand how this > stuff really works. The basic notion is that the decoder reference > picture buffer is considered to be constructed of memory areas that > are structured as frames. When appropriate, two fields will be > placed in the same decoder frame memory storage area. Fields can > only be coupled together for use as a reference frame when this is > the case. So the model basically fits your structuring described in > "(2)", rather than "(1)". But again I caution you to study and > consider all the details before jumping to any overly-general > conclusions. > > Best regards, > > Gary Sullivan > > *From:* mp4-tech-bounces@lists.mpegif.org > [mailto:mp4-tech-bounces@lists.mpegif.org] *On Behalf Of *pankaj > bansal > *Sent:* Tuesday, May 03, 2024 10:12 PM > *To:* mp4-tech@lists.mpegif.org > *Subject:* [Mp4-tech] interlace field support in H.264 > > I have one doubt related to the decoded reference buffer in the > case of interlaced support. Is there only frames are stored or > field can be stored? > for example suppose ref_buffer[8] is a buffer used to store the > decoded ref. frames or fields. Now which of the following thing > can be possible > (1) > ref_buffer[0] store frame1 > ref_buffer[1] store top_field(frame2) > ref_buffer[2] store bottom_feld(frame2) ........... > > (2) > ref_buffer[0] store frame(frame1) > ref_buffer[1] store top_field and bottom_field of (frame2) > > thanks & regards > pankaj > > > __________________________________________________ > 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 > > Yahoo! Mail Mobile > Take Yahoo! Mail with you! > > Check email on your mobile phone. > > > ------------------------------------------------------------------------ > > _______________________________________________ > 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 bharatsoni gmail.com Fri May 6 16:14:13 2005 From: bharatsoni gmail.com (Bharat Soni) Date: Fri May 6 09:30:17 2005 Subject: [Mp4-tech] SAD-PSNR (currrent/reference-original/reconstructed) In-Reply-To: References: Message-ID: Hi Kaustubh, Did you mean original frame is also used as reference picture instead of reconstructed frame on the encoder side? I think the reason behind using reconstructed frame as reference frame on the encoder side is that on the decoder side you have reconstructed (decoded) frame available and not the original one. Of cours it is possible to use original frame as well for the reference. In this case the result might be slightly different. Regards, Bharat On 5/5/05, kaustubh.patankar@vsnl.net wrote: > Hi Bharat > > One correction, there are implementations where original is > also used for encoding. > > with regards > > Kaustubh > > ----- Original Message ----- > From: Bharat Soni > Date: Monday, May 2, 2024 4:58 pm > Subject: Re: [Mp4-tech] SAD-PSNR (currrent/reference-original/reconstructed) > > > Hi Janis, > > I think there is a little bit confusion in which > > frame(original/reconstructed) is used when. > > Lets consider following frames:I"1: Original Frame to be coded as > > I frameI1: is reconstructed (decoded) frame that will be used as > > reference frame.P"2: original frame which has to be predicted from > > I1 and coded as P frame. > > You may (need to) determine the encoding quality of I"1 only when > > ithas already been compressed and again decompressed to > > generatereconstructed frame. So to determine the quality of > > encoding (PSNR),contents (block data) of reconstructed I1 will be > > compared with theoriginal data of I1 i.e. I"1. (I"1 <=> I1) > > For encoding P1, we use the RECONSTRUCTED I1 and NOT the original > > I"1frame as at the decoder side, we do not have the original > > frameavailable for motion compensation.So on the encoder side for > > forming prediction, we will compare thecontents of the current > > (original) frame data with the contents of thereference > > (reconstructed) frame to determine SAD/MAD etc..(P"2 <=> I1). > > Regarding relation between SAD and PSNR, I would say there is > > inverserelation between the two. lower SAD means the blocks are > > verysimilar, this in turn means lower error values of the > > coefficient thatmeans lower the chances of the error due to > > quantization (thisstatement may not be true always, but this is > > how it works). Smallererror means the reconstructed frame will be > > close to original hence itgives high PSNR value, which inturn > > represent good encoding quality. > > I hope its clear now. > > Regards,Bharat > > > > On 4/27/05, janis wrote:> Dear all, > > In > > the previous question about SAD and PSNR, > what I really want to > > know is this, > > To get a SAD, we compare the currnet frame block > > with a reference frame> block. > But to get MSE or PSNR as a > > measurement for the quality, > we compare the reconstructed frame > > with the original frame. > > But there ARE some relation between > > SAD and PSNR, > that when we find the smallest SAD, we can get > > better quality. > > So, why is this happening? > How can I tell > > the relation between comparing 'the original with the> > > reconstructed' > and 'the current with reference'? > > Hope I made > > my question clear, > Thanks in advance, > janis > > > "?? ???, > > Daum" http://www.daum.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 hav! > > e.> > Note: Conduct on the mailing list is subject to the > > Antitrust guidelines> found at> " > > target="l">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 > > From m.bojnordi ece.ut.ac.ir Fri May 6 17:26:55 2005 From: m.bojnordi ece.ut.ac.ir (m.bojnordi@ece.ut.ac.ir) Date: Fri May 6 09:30:22 2005 Subject: [Mp4-tech] MB Modes Message-ID: <17566.217.218.38.17.1115380615.squirrel@217.218.38.17> Hi, Could somebody describe Macroblock modes 0, ..., 6? Regards Mahdi From robin94539 yahoo.com Sat May 7 01:41:22 2005 From: robin94539 yahoo.com (Robin Zoo) Date: Sat May 7 15:38:29 2005 Subject: [Mp4-tech] transcoding, core subfunction supporting multiple standards Message-ID: <20050507074123.66288.qmail@web52001.mail.yahoo.com> Though it is known that H.264 is becoming the main stream of video coding standard, other standards such as MPEG1/2/4, WMV have been and will be in exist in the foreseeable future. As a result, video transcoding is getting more and more important. Could somebody tell me how advanced the technology is, is it worthwhile to build each core subfunction such as IDCT, prediction & compensation or deblocking & deringing filtering supporting multiple video coding standards? So far, I have created a high speed, very compact IDCT(JPEG/MPEG1-2-4/WMV(VC-1)/RealVideo/H.264/AVS(China Std)) engine and a prediction(MPEG4/WMV(VC-1)/RealVideo/H.264/AVS(China Std)) engine. Thanks in advance for your kind advice or comment. Best Regards, Robin __________________________________ Do you Yahoo!? Make Yahoo! your home page http://www.yahoo.com/r/hs From garysull windows.microsoft.com Sat May 7 22:08:16 2005 From: garysull windows.microsoft.com (Gary Sullivan) Date: Sun May 8 08:35:16 2005 Subject: [Mp4-tech] mismatch between the reference software and spec ontransform_size_8x8_flag Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2460E7303C9@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> Tony et al, I don't think there is a problem in either the document or the software. I don't know whether you are looking at a recent draft or not. In the macroblock layer syntax (subclause 7.3.5), transform_size_8x8_flag appears before the coded_block_pattern for I_NxN macroblocks, and after the coded_block_pattern (when present) for other types of macroblocks. If you really think you are seeing something different than that somewhere, please be more specific about where you see a problem. However, my guess is that you were assuming that the order would be the same in all cases, which is not the way it is specified. Best Regards, Gary Sullivan +> -----Original Message----- +> From: mp4-tech-bounces@lists.mpegif.org +> [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Tong Wu +> Sent: Thursday, May 05, 2024 10:43 AM +> To: mp4-tech@lists.mpegif.org +> Subject: [Mp4-tech] mismatch between the reference software +> and spec ontransform_size_8x8_flag +> +> Hi, +> +> In the bistream syntax specification the +> transform_size_8x8_flag syntax is ahead of the +> coded_block_pattern syntax. However when I read the +> readCBPandCoeffsFromNAL()in the reference software +> JM93, I find the coded_block_pattern syntax is parsed +> before transform_size_8x8_flag. +> +> Does anybody notice this mismatch? +> +> Thanks. +> +> Tony +> +> __________________________________________________ +> 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-Ant +> itrust.php +> From w_tong2000 yahoo.com Sat May 7 23:36:57 2005 From: w_tong2000 yahoo.com (Tong Wu) Date: Sun May 8 08:35:21 2005 Subject: [Mp4-tech] mismatch between the reference software and spec ontransform_size_8x8_flag In-Reply-To: 6667 Message-ID: <20050508053657.77202.qmail@web50706.mail.yahoo.com> Hi Gary, I'm looking at the JVT-L012d2.doc (or JVT-K051) and JM94. I can tell that the reference software JM94 does the way as you said and it makes sense. But in the subclause JVT-L012d2.doc as far as I can tell the syntax transform_size_8x8_flag is really always before coded_block_pattern when both are present. Can you please double check it? Thank you very much for your help. Tong --- Gary Sullivan wrote: > > Tony et al, > > I don't think there is a problem in either the > document or the software. > I don't know whether you are looking at a recent > draft or not. > > In the macroblock layer syntax (subclause 7.3.5), > transform_size_8x8_flag appears before the > coded_block_pattern for I_NxN > macroblocks, and after the coded_block_pattern (when > present) for other > types of macroblocks. > > If you really think you are seeing something > different than that > somewhere, please be more specific about where you > see a problem. > However, my guess is that you were assuming that the > order would be the > same in all cases, which is not the way it is > specified. > > Best Regards, > > Gary Sullivan > > +> -----Original Message----- > +> From: mp4-tech-bounces@lists.mpegif.org > +> [mailto:mp4-tech-bounces@lists.mpegif.org] On > Behalf Of Tong Wu > +> Sent: Thursday, May 05, 2024 10:43 AM > +> To: mp4-tech@lists.mpegif.org > +> Subject: [Mp4-tech] mismatch between the > reference software > +> and spec ontransform_size_8x8_flag > +> > +> Hi, > +> > +> In the bistream syntax specification the > +> transform_size_8x8_flag syntax is ahead of the > +> coded_block_pattern syntax. However when I read > the > +> readCBPandCoeffsFromNAL()in the reference > software > +> JM93, I find the coded_block_pattern syntax is > parsed > +> before transform_size_8x8_flag. > +> > +> Does anybody notice this mismatch? > +> > +> Thanks. > +> > +> Tony > +> > +> > __________________________________________________ > +> 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-Ant > +> itrust.php > +> > Yahoo! Mail Stay connected, organized, and protected. Take the tour: http://tour.mail.yahoo.com/mailtour.html From garysull windows.microsoft.com Sat May 7 23:54:22 2005 From: garysull windows.microsoft.com (Gary Sullivan) Date: Sun May 8 08:35:26 2005 Subject: [Mp4-tech] mismatch between the reference software and spec ontransform_size_8x8_flag Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2460E7303D3@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> Tong et al, I've said this many times and I'll probably have to say it again - JVT-L012 was an INPUT document that was never endorsed as an official draft by the JVT. It was also written BEFORE we were finished with FRExt. That document is JUNK. I refuse to double-check anything in that document. Attached is an introduction to the JVT and its work that I sent out last week. See the instructions there about how to get a FREE copy of an OFFICIAL version of the standard that is in much better shape than some random old input document. Best regards, Gary Sullivan +> -----Original Message----- +> From: Tong Wu [mailto:w_tong2000@yahoo.com] +> Sent: Saturday, May 07, 2024 10:37 PM +> To: Gary Sullivan; mp4-tech@lists.mpegif.org +> Subject: RE: [Mp4-tech] mismatch between the reference +> software and spec ontransform_size_8x8_flag +> +> Hi Gary, +> +> I'm looking at the JVT-L012d2.doc (or JVT-K051) and +> JM94. I can tell that the reference software JM94 does +> the way as you said and it makes sense. But in the +> subclause JVT-L012d2.doc as far as I can tell the +> syntax transform_size_8x8_flag is really always before +> coded_block_pattern when both are present. Can you +> please double check it? +> +> Thank you very much for your help. +> +> Tong +> +> --- Gary Sullivan +> wrote: +> > +> > Tony et al, +> > +> > I don't think there is a problem in either the +> > document or the software. +> > I don't know whether you are looking at a recent +> > draft or not. +> > +> > In the macroblock layer syntax (subclause 7.3.5), +> > transform_size_8x8_flag appears before the +> > coded_block_pattern for I_NxN +> > macroblocks, and after the coded_block_pattern (when +> > present) for other +> > types of macroblocks. +> > +> > If you really think you are seeing something +> > different than that +> > somewhere, please be more specific about where you +> > see a problem. +> > However, my guess is that you were assuming that the +> > order would be the +> > same in all cases, which is not the way it is +> > specified. +> > +> > Best Regards, +> > +> > Gary Sullivan +> > +> > +> -----Original Message----- +> > +> From: mp4-tech-bounces@lists.mpegif.org +> > +> [mailto:mp4-tech-bounces@lists.mpegif.org] On +> > Behalf Of Tong Wu +> > +> Sent: Thursday, May 05, 2024 10:43 AM +> > +> To: mp4-tech@lists.mpegif.org +> > +> Subject: [Mp4-tech] mismatch between the +> > reference software +> > +> and spec ontransform_size_8x8_flag +> > +> +> > +> Hi, +> > +> +> > +> In the bistream syntax specification the +> > +> transform_size_8x8_flag syntax is ahead of the +> > +> coded_block_pattern syntax. However when I read +> > the +> > +> readCBPandCoeffsFromNAL()in the reference +> > software +> > +> JM93, I find the coded_block_pattern syntax is +> > parsed +> > +> before transform_size_8x8_flag. +> > +> +> > +> Does anybody notice this mismatch? +> > +> +> > +> Thanks. +> > +> +> > +> Tony +> > +> +> > +> +> > __________________________________________________ +> > +> 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-Ant +> > +> itrust.php +> > +> +> > +> +> +> +> Yahoo! Mail +> Stay connected, organized, and protected. Take the tour: +> http://tour.mail.yahoo.com/mailtour.html +> +> -------------- next part -------------- An embedded message was scrubbed... From: "Gary Sullivan" Subject: RE: Introductory information about the JVT and its work Date: Fri, 6 May 2024 12:16:34 -0700 Size: 7967 Url: /pipermail/mp4-tech/attachments/20050507/0e0ed5cf/attachment.mht From catrauser catrasoftware.it Sun May 8 17:22:52 2005 From: catrauser catrasoftware.it (catrauser) Date: Sun May 8 14:21:05 2005 Subject: [Mp4-tech] The Catra Streaming Platform now open source Message-ID: <20050508142252.29620.qmail@webmaild.fe1.aruba.it> Hi all, it's a pleasure for me to announce the release of the Catra Streaming Platform as an open source project. The Catra Streaming Platform is a 3GPP and ISMA streaming platform able to stream both live and off-line content. You can download it from http://sourceforge.net/projects/openstreaming. You can find also a binary light version of the platform running on Windows just if someone wants to try it without use any development tools like C++ compiler, ... Best regards giu ------------------------------------------------------------------- CatraSoftware Home page: http://www.catrasoftware.it e-mail: catrasoftware-support@catrasoftware.it Streaming page: http://www.catrasoftware.it/Streaming/CatraStreamingPlatform.htm Mailing list: catrastreaming-list@catrasoftware.it e-mail: catrastreaming-support@catrasoftware.it ------------------------------------------------------------------- From minm necas.nec.com.cn Mon May 9 14:59:21 2005 From: minm necas.nec.com.cn (minmin) Date: Mon May 9 03:21:53 2005 Subject: [Mp4-tech][H.264] questions about picture order count Message-ID: <20050509135444.6FCD.MINM@necas.nec.com.cn> Hi Experts! Maybe my question is foolish, but I am not very clear about the relationship among PicOrderCntMsb,pic_order_cnt_lsb and PicOrderCnt described in ISO/IEC 14496-10 clause 8.2.1.1. And what do "msb" and "lsb" exactly mean? If I want to extract only I frames(include IDR and non-IDR) from the bitstream, and form these frames into a new bitstream.Could it be decoded correctly?Is it necesary for each frame to reset some fields(such as "frame_num","pic_order_cnt_lsb" etc.) associated with POC? And is there any other fields needed to reset? And how to reset? Hope for your reply! Thanks a lot! -- minmin From abdessamad.falhi epfl.ch Mon May 9 10:10:28 2005 From: abdessamad.falhi epfl.ch (abdessamad.falhi@epfl.ch) Date: Mon May 9 09:19:34 2005 Subject: [Mp4-tech] documention on aac decoder Message-ID: <1115622628.427f0ce471348@imapwww.epfl.ch> Hi Sakthi, I would like to thank you for your help regards Abdessamad From abdessamad.falhi epfl.ch Mon May 9 15:33:26 2005 From: abdessamad.falhi epfl.ch (abdessamad.falhi@epfl.ch) Date: Mon May 9 09:23:40 2005 Subject: [Mp4-tech] Implementing a simple LC aac decoder Message-ID: <1115642006.427f5896cad7d@imapwww.epfl.ch> Hi, Can someone tell me what are the strict minimum of files that I need to beuild an LC aac decoder. Does anybody have a documentation explaining the blocs of the LC aac decoder and how to implement it? thanks Abdessamad From csa03030 cse.iitd.ernet.in Mon May 9 19:05:13 2005 From: csa03030 cse.iitd.ernet.in (Divyashikha Sethia) Date: Mon May 9 09:26:07 2005 Subject: [Mp4-tech] MPEG4 Base and Enhanced layer streaming Message-ID: <60048.61.246.9.212.1115642113.squirrel@www.cse.iitd.ernet.in> Hello, I am new to MPeg4. With Scalable Encoding Profile in MPEG4, is it possible to send out two streams simultaneously on two different channels, with : i) one with base and enhanced layer. ii) sencond with only the base layer. I want the receiver to have two channels : i) Primary channel with base and enhanced layer ii) Secondary channel with base layer When the Priamary channel goes down the receiver should be able to switch to the Secondary.The video would be in sync with the Primary, since both of them have the same base layer available but with different qualities. Can such a thing be done with MPEG4 ? i need some guidance on this issue. Thanks and Regards, Divya -- Divyashikha Sethia MTech Student Department of Computer Science Indian Institute of Technology, Delhi,India. Ph:91-9350171161 From squ ati.com Mon May 9 10:59:46 2005 From: squ ati.com (Strong Qu) Date: Tue May 10 09:06:17 2005 Subject: [Mp4-tech] [video] worst case of video frame size Message-ID: <44C98333C27112479FFE3E49E636F4F90521F59E@cvexch15.atitech.com> Hi, I'm new on Mpeg4 standard so I hope somebody here can give me some help. I'm developing mpeg4 video encoding application. For simple visual profile level 0 and level 1, how to calculate the maximum size of an encoded frame? According Annex D, there are three models. Are all these models implemented in mpeg4 codec? How can I know is short header used or not? Thanks in advance. - Strong -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050509/c25aa4f2/attachment.html From koen.dewolf ugent.be Mon May 9 17:05:29 2005 From: koen.dewolf ugent.be (Koen De Wolf) Date: Tue May 10 09:10:15 2005 Subject: [Mp4-tech] Motion Estimation In-Reply-To: <20050505095852.44652.qmail@web52608.mail.yahoo.com> Message-ID: <200505091403.j49E34t09395@trappist.elis.UGent.be> Hello Shailendra, There is no such standard way to test the correctness of the ME/MC process. Because of the fact that the ME process is tightly coupled with the rate control algorithm, motion vectors (MV) may differ when bitrates vary. The goal of the ME/MC process together with the rate control is to maximize the quality of the decoded picture by minimizing the bits needed for the transmission of the MB mode decisions, the MVs and the coded prediction errors. Also note that the ME process is not standardized. Depending on the ME algorithm you use, you might even not be able to determine the "ideal" MVs. One way to have an idea about the correctness of your ME/MC is by comparing the motion compensated picture with the original one by means of a matching criterion (such as SAD, SSE, etc). Then, if your ME algorithm is a full search one, there should be no other motion vector field (using the same MB partition modes and search window) that results in a lower SAD, SSE, etc. You might also want to experiment a little with the VCDemo tool (http://ict.ewi.tudelft.nl/index.php?Itemid=124) Hope this helps, Koen -- Multimedia Lab Ghent University - ELIS - IBBT URL http://multimedialab.elis.ugent.be > -----Original Message----- > From: mp4-tech-bounces@lists.mpegif.org > [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of > Shailendra Gupta > Sent: donderdag 5 mei 2005 11:59 > To: mp4-tech@lists.mpegif.org > Subject: [Mp4-tech] Motion Estimation > > Hi; > Please can any one suggest the standard ways to test the > correctness of Motion Estimation and Motion Compensation > vectors just like there are ways for DCT/IDCT, VLCs. > > Regards > Shailendra > > __________________________________________________ > 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-Anti > trust.php > From somdutta.banerjee tcs.com Mon May 9 22:42:08 2005 From: somdutta.banerjee tcs.com (somdutta.banerjee@tcs.com) Date: Tue May 10 09:12:02 2005 Subject: [Mp4-tech] NAL buffering for H.264 Decoder. In-Reply-To: <200505081608.j48G5bpJ028295@lists1.magma.ca> Message-ID: Dear All, In order to implement a H.264 decoder, setting up a buffering logic for the incoming NAL units is essential. However, i dont seem to have any idea as to how long a NAL unit can be at the most ; or whether it is possible to start the decoding process without receiving the entire NAL first. My system also doesnt have enough memory to buffer up entire NALs for some sequences viz. stefan.264, football.264. Pls enlighten me in this regard. Thanks in advance. Somdutta Banerjee Tata Consultancy Services Limited Mailto: somdutta.banerjee@tcs.com Website: http://www.tcs.com Notice: The information contained in this e-mail message and/or attachments to it may contain confidential or privileged information. If you are not the intended recipient, any dissemination, use, review, distribution, printing or copying of the information contained in this e-mail message and/or attachments to it are strictly prohibited. If you have received this communication in error, please notify us by reply e-mail or telephone and immediately and permanently delete the message and any attachments. Thank you -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050509/ea149496/attachment.html From smithajo gmail.com Tue May 10 17:06:52 2005 From: smithajo gmail.com (Smitha Joseph) Date: Tue May 10 09:14:16 2005 Subject: [Mp4-tech] [H.264][Video] Regarding decoding temporal direct luma motion vector Message-ID: <7bca48b405051003366e5a3c59@mail.gmail.com> Hi experts, Thanks for your responses to my earlier queries. I have a doubt regarding the decoding of temporal direct luma motion vector. The section 8.4.1.2.3 of the standard specifies that if vertMvScale is equal to One_To_One and current macroblock is a frame macroblock, MapColToList0( refIdxCol ) returns the lowest valued reference index refIdxL0 in the current reference picture list RefPicList0 that references refPicCol. RefPicList0 shall contain refPicCol. Does this mean that MapColToList0( refIdxCol ) returns refIdxL0 such that RefPicList0[refIdxL0] will be pointing to the same picture as RefPicCol ie., the picnums of the pictures accessed from the decoded picture buffer using RefPicList0[refIdxL0] and RefPicCol will be the same. Or does it mean that MapColToList0( refIdxCol ) returns refIdxL0 such that RefPicList0[refIdxL0] uses picture accessed using RefPicCol from the decoded picture buffer for interprediction. Could you please help me out. Your help will be most appreciated. Thanks, Smitha Joseph From cho8629 dreamwiz.com Tue May 10 23:32:44 2005 From: cho8629 dreamwiz.com (=?ks_c_5601-1987?B?wbbAzciv?=) Date: Tue May 10 11:06:14 2005 Subject: [Mp4-tech] intra prediction mode decision using RD optimization Message-ID: <200505101332.j4ADWss6009091@smtp0.dreamwiz.com> Dear experts.. Well..I thought that in JM, intra prediction mode decision using RD Optimization of luma and chroma may be processed independently. But checking source code in JM 9.2, it is not processed independently between luma and chroma, I think. In JM 9.2, to decide best intra prediction mode of luma and chroma using RD chroma intra prediction is done firstly. And then for each chroma intra prediction mode RDcost is calculated including luma and chroma. Do I understand correctly? If I understand correctly, why does it process like that? Simply thinking about that, best chroma intra prediction mode can be calculated in chroma residual coding?and best luma intra prediction can be done equally. I thinks above both method is just same? Is there anyone who can explain about intra prediction mode decision based on RD? Or I hope to let me know where can I get the document or paper about that? ___________________________________________________________________________ ??? (Ik-Hwan Cho) 402-751 ??? ?? ??3? ????? ????? ????? ???(2N575) 2N575, Multimedia Lab. Electronic Engineering, Inha University Yong Hyun 3 Dong, NamGu, Incheon, 402-751, South Korea E-mail : cho8629@dreamwiz.com MSN : cho8629@korea.com Tel : +82-32-860-7415 Mobile : +82-018-267-4535 __________________________________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050510/18b7378b/attachment.html From jc sj.co.uk Tue May 10 15:44:15 2005 From: jc sj.co.uk (John Cox) Date: Tue May 10 11:06:20 2005 Subject: [Mp4-tech] NAL buffering for H.264 Decoder. In-Reply-To: References: <200505081608.j48G5bpJ028295@lists1.magma.ca> Message-ID: Hi >Dear All, > In order to implement a H.264 decoder, setting up a buffering >logic for the incoming NAL units is essential. > > However, i dont seem to have any idea as to how long a NAL unit >can be at the most ; or whether it is possible to start the decoding >process without receiving the entire NAL first. Worst case NAL unit would seem to be (size of a raw frame / 2) where 2 is the Min Compression Ratio from Annex A, so in the case of 4:2:0 D1 that would be 311,040 bytes, though it is likely to be smaller than that due to bit rate restrictions. My decoder will take data in ~512 byte chunks (worst case MB + a bit), but requires much, much more than that by way of intermediate storage and reference buffers. I've no idea what yours requires. > My system also doesnt have enough memory to buffer up entire NALs >for some sequences viz. stefan.264, football.264. If you don't have enough memory to buffer a frame of compressed data then you are probably in real trouble with the uncompressed data & reference buffers. > Pls enlighten me in this regard. > >Thanks in advance. > >Somdutta Banerjee >Tata Consultancy Services Limited >Mailto: somdutta.banerjee@tcs.com >Website: http://www.tcs.com John Cox SJ Consulting From liat_appel yahoo.com Tue May 10 08:38:00 2005 From: liat_appel yahoo.com (liat appel) Date: Tue May 10 11:06:26 2005 Subject: [Mp4-tech] mpeg4 parser Message-ID: <20050510143800.50140.qmail@web40910.mail.yahoo.com> Hi, I need to write a parser for MPEG4 and H263 bit streams, to signal when there is a VOP, VOL, GOV header. Is there an available tool or a free code? Thanks, Liat. Discover Yahoo! Use Yahoo! to plan a weekend, have fun online and more. Check it out! http://discover.yahoo.com/ From tomotohara yahoo.com Tue May 10 10:54:25 2005 From: tomotohara yahoo.com (Tomo) Date: Wed May 11 09:30:30 2005 Subject: [Mp4-tech] NAL buffering for H.264 Decoder. In-Reply-To: Message-ID: <20050510165425.14942.qmail@web50907.mail.yahoo.com> Isn't the relating information found in the table of "level" ? --- somdutta.banerjee@tcs.com wrote: > > Dear All, > In order to implement a H.264 decoder, setting up a buffering > > logic for the incoming NAL units is essential. > > However, i dont seem to have any idea as to how long a NAL > unit > can be at the most ; or whether it is possible to start the decoding > process without receiving the entire NAL first. > > My system also doesnt have enough memory to buffer up entire > NALs > for some sequences viz. stefan.264, football.264. > > Pls enlighten me in this regard. > > Thanks in advance. > > Somdutta Banerjee > Tata Consultancy Services Limited > Mailto: somdutta.banerjee@tcs.com > Website: http://www.tcs.com > > Notice: The information contained in this e-mail message and/or > attachments to it may contain confidential or privileged information. > If you are not the intended recipient, any dissemination, use, > review, distribution, printing or copying of the information > contained in this e-mail message and/or attachments to it are > strictly prohibited. If you have received this communication in > error, please notify us by reply e-mail or telephone and immediately > and permanently delete the message and any attachments. Thank you> _______________________________________________ > 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 Tomo From hworlddoc gmail.com Wed May 11 09:21:04 2005 From: hworlddoc gmail.com (Jason Huang) Date: Wed May 11 09:30:35 2005 Subject: [Mp4-tech] [H.264][Video] Regarding decoding temporal direct luma motion vector In-Reply-To: <7bca48b405051003366e5a3c59@mail.gmail.com> References: <7bca48b405051003366e5a3c59@mail.gmail.com> Message-ID: <4bb4853f05051017215a69023b@mail.gmail.com> Are you familiar to MPEG-4? If yes, try to imagine how the direct mode in these standards works, then you will understand. Xiaojie Huang On 5/10/05, Smitha Joseph wrote: > Hi experts, > > Thanks for your responses to my earlier queries. I have a doubt > regarding the decoding of temporal direct luma motion vector. > > The section 8.4.1.2.3 of the standard specifies that if vertMvScale is > equal to One_To_One and current macroblock is a frame macroblock, > MapColToList0( refIdxCol ) returns the lowest valued reference index > refIdxL0 in the current reference picture list RefPicList0 that > references refPicCol. RefPicList0 shall contain refPicCol. > > Does this mean that MapColToList0( refIdxCol ) returns refIdxL0 such > that RefPicList0[refIdxL0] will be pointing to the same picture as > RefPicCol ie., the picnums of the pictures accessed from the decoded > picture buffer using RefPicList0[refIdxL0] and RefPicCol will be the > same. > > Or does it mean that MapColToList0( refIdxCol ) returns refIdxL0 such > that RefPicList0[refIdxL0] uses picture accessed using RefPicCol from > the decoded picture buffer for interprediction. > > Could you please help me out. Your help will be most appreciated. > > Thanks, > Smitha Joseph > > _______________________________________________ > 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 tariqhusain rediffmail.com Wed May 11 02:08:15 2005 From: tariqhusain rediffmail.com (tariq husain shaikh) Date: Wed May 11 09:30:41 2005 Subject: [Mp4-tech] [AUDIO]- AAC test vectors Message-ID: <20050511010921.13881.qmail@webmail31.rediffmail.com> ? hello, I am trying to download the ADIf and ADTS version of the ct_psytel test vector for AAC. I have tried the link mentioned below, but cant download. http://home.arcor.de/hans-juergen.bardenhagen/ct_psytel-adif.aac Is the links still working. Can someone please suggest any alternate links for the same. Thanks in advance Tariq -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050511/46499ad8/attachment.html From mpeg2_user yahoo.com Tue May 10 20:07:58 2005 From: mpeg2_user yahoo.com (Tom P) Date: Wed May 11 09:30:47 2005 Subject: [Mp4-tech] Concepts of FMO and ASO Message-ID: <20050511020758.22180.qmail@web54705.mail.yahoo.com> HI All, Can somebody explain me the concepts of Flexible Macroblock Ordering(FMO) and Arbitrary Slice Ordering(ASO) of H.264 ? Thanks in advance, --------------------------------- Yahoo! Mail Mobile Take Yahoo! Mail with you! Check email on your mobile phone. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050510/e650b7db/attachment.html From bharatsoni gmail.com Wed May 11 10:16:21 2005 From: bharatsoni gmail.com (Bharat Soni) Date: Wed May 11 09:30:52 2005 Subject: [Mp4-tech] mpeg4 parser In-Reply-To: <20050510143800.50140.qmail@web40910.mail.yahoo.com> References: <20050510143800.50140.qmail@web40910.mail.yahoo.com> Message-ID: Hi, I think there is a freeware available at www.venera.co.in . -Bharat On 5/10/05, liat appel wrote: > Hi, > I need to write a parser for MPEG4 and H263 bit > streams, to signal when there is a VOP, VOL, GOV > header. > Is there an available tool or a free code? > Thanks, > Liat. > > Discover Yahoo! > Use Yahoo! to plan a weekend, have fun online and more. Check it out! > http://discover.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 v_v_bhat yahoo.com Tue May 10 22:32:13 2005 From: v_v_bhat yahoo.com (vinayak bhat) Date: Wed May 11 09:30:57 2005 Subject: [Mp4-tech] Regarding H.263 encoder Message-ID: <20050511043213.16739.qmail@web41603.mail.yahoo.com> Hi all, In case of H.263 encoder if Annex T is made On...how does it improve the quality /performance...? For the H.263 Encoder if Annex T support is added how do we test weather its working or know..is there any tool ..i mean H.263 Encoder stream analysis tool. thanks --vinayak __________________________________________________ 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/20050510/37f6d3cc/attachment.html From gbmallikarjunarao yahoo.com Wed May 11 07:00:37 2005 From: gbmallikarjunarao yahoo.com (mallikarjun rao) Date: Wed May 11 09:31:03 2005 Subject: [Mp4-tech] [audio][MPEG4] Conformance Testing of HE- AAC Decoder In-Reply-To: 6667 Message-ID: <20050511050037.57196.qmail@web52004.mail.yahoo.com> Hi, can anybody tell me the HE- AAC Decoder testing procedure? I downloaded sbr test vectors from ftp://mpaudconf:adif2mp4@ftp.iis.fraunhofer.de/ . what is the maximum error limit for those ref. test vectors compared to our decoder output? can you please send related information or web addresses? Thanks in advance Mallikarjun --------------------------------- How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos. Get Yahoo! Photos -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050511/52e814f0/attachment.html From narenlj sasken.com Wed May 11 12:45:43 2005 From: narenlj sasken.com (Narendra Joshi) Date: Wed May 11 09:31:08 2005 Subject: [Mp4-tech][H264 video] - question about temporal and spatial prediction mode Message-ID: Hello there, I am trying to understand Inter prediction process for B_skip and B_direct macroblocks. >From the information mentioned in section 8.4.1.2, the difference between temporal and spatial prediction mode is not clear to me. Could someone explain to me the conceptual difference between temporal and spatial prediction modes in layman's language ? Also, it's not clear as to why RefPicList1 is used always (and not RefPicList0) to pick up co-located macroblock / picture. (am referring to Table 8.6). Please help ! Thanks & regards, Narendra "SASKEN RATED THE BEST EMPLOYER IN THE COUNTRY by the BUSINESS TODAY Mercer Survey 2004" SASKEN BUSINESS DISCLAIMER This message may contain confidential, proprietary or legally Privileged information. In case you are not the original intended Recipient of the message, you must not, directly or indirectly, use, Disclose, distribute, print, or copy any part of this message and you are requested to delete it and inform the sender. Any views expressed in this message are those of the individual sender unless otherwise stated. Nothing contained in this message shall be construed as an offer or acceptance of any offer by Sasken Communication Technologies Limited ("Sasken") unless sent with that express intent and with due authority of Sasken. Sasken has taken enough precautions to prevent the spread of viruses. However the company accepts no liability for any damage caused by any virus transmitted by this email -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050511/a673a41d/attachment.html From ernie_e_lim yahoo.com.au Wed May 11 17:29:43 2005 From: ernie_e_lim yahoo.com.au (Ernest Lim) Date: Wed May 11 09:31:14 2005 Subject: [Mp4-tech] H264 DPB size for practical decoders Message-ID: <20050511062943.57908.qmail@web53505.mail.yahoo.com> Hello all, I have a question regarding the size of the DPB and how this may be affected by the POC of decoded frames. I mailed this to JVT-EXPERTS but got no reply. Presumably on a practical system, possibly real-time, where for every period in which a frame is decoded, a frame is also displayed. Then if POC is not ordered in the same manner as frame number, there has to be a delay after the decoding commences before frames can be displayed. As I understand, this delay is determined by num_reorder_frames, which has a default value of 16 and is specified in the HRD parameter set. Do the decoded frames that are awaiting display, that are not being used for reference, reside in the DPB? This is what seems to happen in JM, although JM is able to "display" many frames in a single period so that when the DPB is filled with these frames awaiting display and reference frames, space is created by displaying and removing the next non-reference frame according to POC. What is supposed to happen if only 1 frame can be removed in a decoding period? My feeling is that the DPB should in fact be twice the size in this situation. To illustrate this, consider the following example. In this case, the DPB size is 16, num_reorder_frames is 16 and num_ref_frames is 16. The POCs of the following frames, listed in decoding order is: 0,2,...,15,16,17,1,18,19,...,31,32 The frame with POC 1 has 16 frames that precede it in decode order (2,...,17) but come after it in display order - this is consistent with num_reorder_frames of 16. Suppose that the 16 frames 0,1,2,...,15 are all used as reference frames for the entirety of the decode process once all are decoded and available. We would get the following sequence: 0(.),2(.),...,15(.),16(.),17(0),1(1),18(2),19(3),...,31(15),32(16) where the first number is the POC of the decoded frame and the one in brackets is the POC of the displayed frame. Since we are displaying 1 frame every decode period, the first frame that can be removed from the DPB is after the one with POC of 16 has been displayed. None of 0,1,...,15 can be removed since they are all being used as reference. POC 16 can be removed since it has been displayed and is not being used for reference. However, by this stage we have frames with POCs 0,...,31 in the DPB and are presumably about to insert the most recently decoded frame POC 32 into it! Is this situation correct, or have I misinterpreted something? It suggests that practical decoders that wish to display one frame every period in the correct display order will need a DPB of at least 2 times that stated in the spec. Sorry about the lengthy message. Regards, Ernie. --------------------------------- Find local movie times and trailers on Yahoo! Movies. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050511/2387d379/attachment-0001.html From liat_appel yahoo.com Wed May 11 01:38:30 2005 From: liat_appel yahoo.com (liat appel) Date: Wed May 11 09:31:19 2005 Subject: [Mp4-tech] mpeg4 parser In-Reply-To: 6667 Message-ID: <20050511073830.92519.qmail@web40905.mail.yahoo.com> Thanks. Is there any available source code? Liat. --- Bharat Soni wrote: > Hi, > > I think there is a freeware available at > www.venera.co.in . > > -Bharat > > On 5/10/05, liat appel wrote: > > Hi, > > I need to write a parser for MPEG4 and H263 bit > > streams, to signal when there is a VOP, VOL, GOV > > header. > > Is there an available tool or a free code? > > Thanks, > > Liat. > > > > Discover Yahoo! > > Use Yahoo! to plan a weekend, have fun online and > more. Check it out! > > http://discover.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 > > > __________________________________ Yahoo! Mail Mobile Take Yahoo! Mail with you! Check email on your mobile phone. http://mobile.yahoo.com/learn/mail From Shailendra.Singh skyworksinc.com Wed May 11 16:45:25 2005 From: Shailendra.Singh skyworksinc.com (Shailendra.Singh@skyworksinc.com) Date: Wed May 11 09:31:26 2005 Subject: [Mp4-tech] [video] Format Conversion In-Reply-To: <91D7F2CEE3425A4A9D11311D09FCE2460CC383AE@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> Message-ID: Hi All, Is there a straightforward method for converting a .mov (QuickTime) file (video track only) to a .mp4 file ? Is FFMPEG useful in this case ? Regards, Shailendra -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050511/620e454d/attachment.html From mp3.aac.mp4 gmail.com Wed May 11 20:49:35 2005 From: mp3.aac.mp4 gmail.com (tech list) Date: Wed May 11 10:47:50 2005 Subject: [Mp4-tech] mpeg4 parser In-Reply-To: <20050510143800.50140.qmail@web40910.mail.yahoo.com> References: <20050510143800.50140.qmail@web40910.mail.yahoo.com> Message-ID: <409a09b90505110719541c703f@mail.gmail.com> FAAD has a MP4 file format parser... (audiocoding.com ) On 5/10/05, liat appel wrote: > > Hi, > I need to write a parser for MPEG4 and H263 bit > streams, to signal when there is a VOP, VOL, GOV > header. > Is there an available tool or a free code? > Thanks, > Liat. > > Discover Yahoo! > Use Yahoo! to plan a weekend, have fun online and more. Check it out! > http://discover.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 > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050511/35542a00/attachment.html From alexismt comcast.net Wed May 11 09:47:31 2005 From: alexismt comcast.net (Alexis Michael Tourapis) Date: Thu May 12 08:39:58 2005 Subject: [Mp4-tech][H264 video] - question about temporal and spatial prediction mode In-Reply-To: Message-ID: <200505111547.j4BFlc0f000063@lists1.magma.ca> Dear Narendra, Try first checking some publications that already exist with regards to this topic. i.e. you can try and check the paper published in CSVT on this topic; http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=1377367&isnumber=300 64 For an online version, if you do not have access to ieeexplore or CSVT, although this might not be the final version, you can check at this link http://research.microsoft.com/~fengwu/papers/direct_csvt_04.pdf. Simply speaking, temporal direct uses temporal correlation to derive motion vectors, while spatial direct uses spatial correlation (i.e. one considers mvs from co-located positions in previous pictures, while the second considers mvs from adjacent positions in the same slice). There are also several old contributions to the standard (if I recall correctly these should be for the Klagenfurt meeting in 2002 from either Microsoft or Motorola) so maybe you could try and find those online. With regards to the use of RefPicList1, although this might be a bit reminiscent of older standards, it still relates to the most common coding structures that one may use (i.e. IBBPBBP.). Since the default reference picture order is based on poc directions and distances, the idea would be to also use the picture which could provide the "best" motion vectors for interpolation. Based on the default reference order, pictures in the List0 buffer would mainly (although not necessarily) be coming from the past, and therefore most likely referring to pictures in the past as well, while it is likely (but again not necessarily) that the ones in List1 are coming from the future. However List1 references may still be referring themselves to some of the pictures in List0. As you may observe from this description, the use of List1 is partly made due to the fact that interpolation is usually much more reliable than extrapolation which may be more often required if List0 was used instead. I hope this helps. Alexis _____ From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Narendra Joshi Sent: Tuesday, May 10, 2024 11:16 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech][H264 video] - question about temporal and spatial prediction mode Hello there, I am trying to understand Inter prediction process for B_skip and B_direct macroblocks. >From the information mentioned in section 8.4.1.2, the difference between temporal and spatial prediction mode is not clear to me. Could someone explain to me the conceptual difference between temporal and spatial prediction modes in layman's language ? Also, it's not clear as to why RefPicList1 is used always (and not RefPicList0) to pick up co-located macroblock / picture. (am referring to Table 8.6). Please help ! Thanks & regards, Narendra "SASKEN RATED THE BEST EMPLOYER IN THE COUNTRY by the BUSINESS TODAY Mercer Survey 2004" SASKEN BUSINESS DISCLAIMER This message may contain confidential, proprietary or legally Privileged information. In case you are not the original intended Recipient of the message, you must not, directly or indirectly, use, Disclose, distribute, print, or copy any part of this message and you are requested to delete it and inform the sender. Any views expressed in this message are those of the individual sender unless otherwise stated. Nothing contained in this message shall be construed as an offer or acceptance of any offer by Sasken Communication Technologies Limited ("Sasken") unless sent with that express intent and with due authority of Sasken. Sasken has taken enough precautions to prevent the spread of viruses. However the company accepts no liability for any damage caused by any virus transmitted by this email -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050511/78dbbf2d/attachment.html From mcp24 rediffmail.com Thu May 12 05:18:01 2005 From: mcp24 rediffmail.com (Manjunatha Pawate) Date: Thu May 12 08:40:04 2005 Subject: [Mp4-tech] require good links to understand video compression Message-ID: <20050512041920.13895.qmail@webmail18.rediffmail.com> ? Hi, I am new to mpeg, I want to learn mpeg4.Please let me know any good links for begineers. awaiting for the positive reply, thanks and regards,Manjunath Manjunatha C.Pawate Gangavathi silk saree centre Binny company road Davangere - 577001 Karnataka INDIA ph: 94480 41500 -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050512/8bf010a6/attachment.html From snd codingtechnologies.com Thu May 12 20:17:28 2005 From: snd codingtechnologies.com (Andreas Schneider) Date: Thu May 12 15:40:20 2005 Subject: [Mp4-tech] [audio][MPEG4] Conformance Testing of HE- AAC Decoder In-Reply-To: <20050511050037.57196.qmail@web52004.mail.yahoo.com> Message-ID: Mallikarjun, for proper HE-AAC conformance testing you need a special tool. It is available as part of the audio conformance tools here: ftp://mpaudconf:adif2mp4@ftp.iis.fraunhofer.de/mpeg4audio-conformance/confTools/mpeg4-audio-confTools_strip.zip in the subdirectory conf_sbr. Unfortunately the version that's available here is from some time last year, when changes in other modules broke the SBR conformance_tool. It is possible to get the conformance tool running by changing a few lines in the code: In conf_sbr/makefile, remove * 'drc.o' from the list of OBJS * '-DDRC' from In conf_sbr/src/mp4dec.c, change the line that says: " infoinit(mc_info.sampling_rate_idx); " to " huffbookinit(mc_info.sampling_rate_idx); " and in conf_sbr/src/ decoderConfTool.c remove all lines from the one saying " if (!ready_bitBuffer_sim) { " upto and including " check_bitBuffer_sim(); " This will cripple the tool a bit, but all required functionality will still be there. I hope some time in the near future the snapshot of the conformance tools will be updated so this won't be required anymore. Regards, Andreas mp4-tech-bounces@lists.mpegif.org wrote on 2024-05-11 07:00:37: > Hi, > > can anybody tell me the HE- AAC Decoder testing procedure? > > I downloaded sbr test vectors from ftp://mpaudconf:adif2mp4@ftp.iis. > fraunhofer.de/ . > > what is the maximum error limit for those ref. test vectors compared > to our decoder output? > > can you please send related information or web addresses? > > Thanks in advance > Mallikarjun > > > > > > How much free photo storage do you get? Store your holiday snaps for > FREE with Yahoo! Photos. Get Yahoo! Photos > _______________________________________________ > 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 spsatendra gmail.com Fri May 13 10:04:29 2005 From: spsatendra gmail.com (Satendra) Date: Fri May 13 09:36:45 2005 Subject: [Mp4-tech] Concepts of FMO and ASO In-Reply-To: <20050511020758.22180.qmail@web54705.mail.yahoo.com> References: <20050511020758.22180.qmail@web54705.mail.yahoo.com> Message-ID: <5942723205051220343e322958@mail.gmail.com> Hi, here I am posting some of my notes about FMO and ASO. hopes that it will be useful Flexible Macroblock Ordering (FMO) Usually MBs are assigned to slices in the scan order. But when FMO is enabled, then a slice group map is defined, and there MBs can be assigned to slices in an order other then scan order. FMO is supposed to enhance the error resilience and concealment in the way that, if packet containing one slice data is lost, then it can be interpolated (predicted) by other slices in the picture. For instance if we assign all even MBs to slice 0 and odd MBs to slice 1, then the lost data of any slice can be concealed to some amount by interpolating the MB in lost slice by the four surrounding MBs in the received slice. Although FMO is useful in error prone environments, but it comes with the burden of complexity and reduced coding efficiency. As, prediction across slice boundaries is restricted, so contiguous MBs in the picture if assigned to different slices can't be predicted from each other for the encoding. Arbitrary Slice Ordering (ASO) ASO is the flexibility to receive the slices in any order at the decoder. It is based on the fact that the slices in the H.264 are independent of each other so they can be decoded independently of the other slices in the picture, even if they are arriving at the decoder out of order. This feature can improve end-to-end delay in real time applications, with networks having out of order delivery behavior. On 5/11/05, Tom P wrote: > > HI All, > Can somebody explain me the concepts of Flexible Macroblock Ordering(FMO) > and Arbitrary Slice Ordering(ASO) of H.264 ? > Thanks in advance, > > ------------------------------ > Yahoo! Mail Mobile > Take Yahoo! Mail with you!Check email on your mobile phone. > > > _______________________________________________ > 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 > > -- ----------------------------------------------------------------------------------------------------------------------------------- "We all agree on the necessity of compromise. We just can't agree on when it's necessary to compromise." ------Larry Wall ----------------------------------------------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050513/21537a8e/attachment.html From dattagurubn yahoo.com Thu May 12 23:24:56 2005 From: dattagurubn yahoo.com (Dattaguru B.N.) Date: Fri May 13 09:36:50 2005 Subject: [Mp4-tech] [audio][MPEG4]SBR HF generation and Adjustment In-Reply-To: 6667 Message-ID: <20050513052456.54051.qmail@web20126.mail.yahoo.com> Dear Experts, Can anybody please let me know, 1. What informations of low frequency are required for HF generation and adjustment? 2. What parameters during encoding decide the High frequency generation and adjustment? 3. If I have an AAC bit stream, can I encode the same using SBR encoder? In that case, what will be difference between AAC bit stream compared to same bit stream encoded with SBR? 4. Whose audio quality is better between normal AAC when compared with SBR and why? Thanks in advance, Warm regards, Dattaguru Yahoo! Mail Stay connected, organized, and protected. Take the tour: http://tour.mail.yahoo.com/mailtour.html From harish.vasudeva amd.com Sun May 15 21:29:12 2005 From: harish.vasudeva amd.com (Vasudeva, Harish) Date: Mon May 16 09:26:00 2005 Subject: [Mp4-tech] Huffman table lookup Message-ID: <4A77E4BC05A4EF4D9DAA4C8CB5EEE51CBA32D4@SAUSEXMB1.amd.com> Hi Folks, I am trying to optimize the VLD lookup since its runs extremely slow on Linux (I have Ver 2.3 of the MS MPEG-4 decoder). Just wanted to know if there are any details on what is the minimum requirement for supporting Advanced Simple Profile Progressive content decoding only. Also, I would really appreciate if you could send me a link to any optimized open-source algorithms. Best Regards, HARISH V *A closed mind is a good thing to lose* -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050515/bf11e55a/attachment.html From kjk mcubeworks.com Mon May 16 11:57:31 2005 From: kjk mcubeworks.com (JK(Jeong Kwon) Kim) Date: Mon May 16 09:29:33 2005 Subject: [Mp4-tech] profile_and_level_indication of MPEG4 video ASP In-Reply-To: <20050513052456.54051.qmail@web20126.mail.yahoo.com> Message-ID: <200505160159.j4G1xQcP002443@localhost.localdomain> Dear experts, I have a question about "profile_and_level_indication" in MPEG4 part2. Now, ASP (Advanced Simple Profile) is widely used in our life. So, profile_and_level_indication for ASP shall be defined. (Just as profile_and_level_indication for SP/L0 is defined as '8') However, I can not find ASP in MPEG4 standard. Please let me know about it. Thanks in advance! Regards, JK Kim From loloo 21cn.com Mon May 16 14:47:42 2005 From: loloo 21cn.com (loloo) Date: Mon May 16 09:31:53 2005 Subject: [Mp4-tech] About the coding of ref_idx_l0 and ref_idx_l1 Message-ID: <009501c559da$c8150310$516360ac@EAT> Dear all: I have a question.When we write MV data to bitstream, we use 'mvd' instead of 'mv' to save bits, then when we write ref data, why we write ref_idx_l0(te) and ref_idx_l1(te) directly and not use the predicted method just like 'mvd',for example,we can use 'ref delta'? Regards RuiLu -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050516/a743817b/attachment.html From zombiyaga yahoo.com Mon May 16 00:17:25 2005 From: zombiyaga yahoo.com (Alex) Date: Mon May 16 09:31:59 2005 Subject: [Mp4-tech] [video] mime types Message-ID: <20050516061725.74461.qmail@web53910.mail.yahoo.com> Hi All, is there any document specifying the syntax for video mime types (e. g. H.263 / MPEG4, uppercase / lowercase, etc.) ? I was able to find RFC 3555 - MIME Type Registration of RTP Payload Formats, but it seems that's not actually what I'm looking for. Thanks, -- Regards, Alex --------------------------------- Discover Yahoo! Have fun online with music videos, cool games, IM & more. Check it out! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050515/15d812ce/attachment.html From tariqhusain rediffmail.com Mon May 16 08:05:40 2005 From: tariqhusain rediffmail.com (tariq husain shaikh) Date: Mon May 16 09:32:05 2005 Subject: [Mp4-tech] [AUDIO] - EAAC+ Message-ID: <20050516070700.24000.qmail@webmail9.rediffmail.com> Hello everybody, Is it possible to add ADIF and ADTS file format support to the 3gpp EAAC+ encoder and decoder. Is it OK if i do so...or are there any licensing and copyright issues. If yes, how can i go about doing it. Is there any place to obtain EAAC+ encoded files on the net? Thank you all in advance Regards Tariqhusain -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050516/b534927e/attachment.html From g.lakshmankumar gmail.com Mon May 16 21:26:05 2005 From: g.lakshmankumar gmail.com (Lakshman Kumar) Date: Mon May 16 16:46:13 2005 Subject: [Mp4-tech] Decoding POC Message-ID: <138aa52f05051607562e6879ba@mail.gmail.com> hi all in h.264 decoder POC mode 0 and 1 r represented which calculated MSB and LSB i cant understand why its done so. its available in function decode_poc(); also how DCT separated Low and High freq components -- lakshman kumar From tma iis.fhg.de Mon May 16 21:20:54 2005 From: tma iis.fhg.de (Herbert Thoma) Date: Mon May 16 16:46:18 2005 Subject: [Mp4-tech] profile_and_level_indication of MPEG4 video ASP In-Reply-To: <200505160159.j4G1xQcP002443@localhost.localdomain> References: <200505160159.j4G1xQcP002443@localhost.localdomain> Message-ID: <4288E486.3040408@iis.fhg.de> JK(Jeong Kwon) Kim wrote: > Dear experts, > > I have a question about "profile_and_level_indication" in MPEG4 part2. > Now, ASP (Advanced Simple Profile) is widely used in our life. > So, profile_and_level_indication for ASP shall be defined. > (Just as profile_and_level_indication for SP/L0 is defined as '8') > > However, I can not find ASP in MPEG4 standard. > Please let me know about it. ASP L0: 240 ASP L1: 241 ASP L2: 242 ASP L3: 243 ASP L3b: 247 ASP L4: 244 ASP L5: 245 Table G-1 of ISO/IEC 14496-2:2004 Regards, Herbert. > Thanks in advance! > > Regards, > JK Kim > > _______________________________________________ > 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 Group Manager Video 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 singer apple.com Mon May 16 13:46:39 2005 From: singer apple.com (Dave Singer) Date: Tue May 17 02:34:32 2005 Subject: [Mp4-tech] [video] mime types In-Reply-To: <20050516061725.74461.qmail@web53910.mail.yahoo.com> References: <20050516061725.74461.qmail@web53910.mail.yahoo.com> Message-ID: At 23:17 -0700 15/05/05, Alex wrote: >Hi All, > >is there any document specifying the syntax for video mime types (e. >g. H.263 / MPEG4, uppercase / lowercase, etc.) ? I was able to find >RFC 3555 - MIME Type Registration of RTP Payload Formats, but it >seems that's not actually what I'm looking for. > the MP4 file format registration is still, um, in process. It's supposed to be video/mp4, or for files with no visual aspect at all, audio/mp4 may be used also. >Thanks, > > > > >-- >Regards, >Alex > > > >Discover Yahoo! >Have fun online with music videos, cool games, IM & more. >Check >it out! > >_______________________________________________ >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/20050516/b84fce39/attachment.html From sanjay_gims rediffmail.com Tue May 17 09:12:48 2005 From: sanjay_gims rediffmail.com (guntupalli mrine sanjay) Date: Tue May 17 06:52:14 2005 Subject: [Mp4-tech] MPEG4 Encoder n Decoder source codes Message-ID: <20050517081407.29778.qmail@webmail30.rediffmail.com> ? Hi, So far i seen so many links. But i am not able to know which is use ful for me. Any links or codes, please help me out. Thanks a lot, Sanjay -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050517/5703a47d/attachment.html From lakshman97 yahoo.com Tue May 17 02:15:30 2005 From: lakshman97 yahoo.com (Lakshman) Date: Tue May 17 06:56:17 2005 Subject: [Mp4-tech] [video] mime types In-Reply-To: Message-ID: <20050517081531.23392.qmail@web30213.mail.mud.yahoo.com> Hello all, I used the fallowing MIME types in my project, this is what I found in RFC manuals. MPEG4 Mime Type = "video/MP4V-ES" H263 Mime Type = "video/H263-2000" H264 Mime Type = "video/H264" Thanks and regards -Lakshman --- Dave Singer wrote: > At 23:17 -0700 15/05/05, Alex wrote: > >Hi All, > > > >is there any document specifying the syntax for > video mime types (e. > >g. H.263 / MPEG4, uppercase / lowercase, etc.) ? I > was able to find > >RFC 3555 - MIME Type Registration of RTP Payload > Formats, but it > >seems that's not actually what I'm looking for. > > > > the MP4 file format registration is still, um, in > process. It's > supposed to be video/mp4, or for files with no > visual aspect at all, > audio/mp4 may be used also. > > >Thanks, > > > > > > > > > >-- > >Regards, > >Alex > > > > > > > >Discover Yahoo! > >Have fun online with music videos, cool games, IM & > more. > >Check > > >it out! > > > >_______________________________________________ > >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> _______________________________________________ > 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 __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new resources site! http://smallbusiness.yahoo.com/resources/ From zombiyaga yahoo.com Tue May 17 03:00:14 2005 From: zombiyaga yahoo.com (Alex) Date: Tue May 17 06:58:46 2005 Subject: [Mp4-tech] [video] mime types In-Reply-To: 6667 Message-ID: <20050517090014.62207.qmail@web53909.mail.yahoo.com> Thanks to All, How about the profile / level definition ? e.g. H263: "video/H263-2000; profile=0; level=30" MP4: "video/MP4V-ES; profile-level-id=1" or "video/mp4v-es; profile-level-id=1" ? who is currently working with MIME syntax standartization? It still remains up to developer, how this feature is implemented; Lakshman wrote: Hello all, I used the fallowing MIME types in my project, this is what I found in RFC manuals. MPEG4 Mime Type = "video/MP4V-ES" H263 Mime Type = "video/H263-2000" H264 Mime Type = "video/H264" Thanks and regards -Lakshman --- Dave Singer wrote: > At 23:17 -0700 15/05/05, Alex wrote: > >Hi All, > > > >is there any document specifying the syntax for > video mime types (e. > >g. H.263 / MPEG4, uppercase / lowercase, etc.) ? I > was able to find > >RFC 3555 - MIME Type Registration of RTP Payload > Formats, but it > >seems that's not actually what I'm looking for. > > > > the MP4 file format registration is still, um, in > process. It's > supposed to be video/mp4, or for files with no > visual aspect at all, > audio/mp4 may be used also. > > >Thanks, > > > > > > > > > >-- > >Regards, > >Alex > > > > > > > >Discover Yahoo! > >Have fun online with music videos, cool games, IM & > more. > >Check > > >it out! > > > >_______________________________________________ > >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> _______________________________________________ > 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 __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new resources site! http://smallbusiness.yahoo.com/resources/ -- Regards, Alex --------------------------------- Yahoo! Mail Stay connected, organized, and protected. Take the tour -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050517/4eb1ef14/attachment.html From gripened gmail.com Tue May 17 20:00:29 2005 From: gripened gmail.com (Jayant Chauhan) Date: Tue May 17 07:02:21 2005 Subject: [Mp4-tech] H264 Details Message-ID: <5b996acb05051703002f3d4bbf@mail.gmail.com> Hey all Could anyone please tell me as to how to get 2 things out of the H264 stream at each slice level. One being the Start address for the Parser Top to start (that is, the original location of the bitstream). And the second being the Bit offset for the Parser top to decode within the Start Address. I have using the JM95 decoder, but something is going wrong, as I have tweaked with the code, as in, the bit offset and the start address and size for each slice is coming out wrong, with regards Jayant From singer apple.com Tue May 17 05:55:54 2005 From: singer apple.com (Dave Singer) Date: Tue May 17 09:31:57 2005 Subject: [Mp4-tech] [video] mime types In-Reply-To: <20050517081531.23392.qmail@web30213.mail.mud.yahoo.com> References: <20050517081531.23392.qmail@web30213.mail.mud.yahoo.com> Message-ID: At 1:15 -0700 17/05/05, Lakshman wrote: >Hello all, > >I used the fallowing MIME types in my project, this is >what I found in RFC manuals. > >MPEG4 Mime Type = "video/MP4V-ES" >H263 Mime Type = "video/H263-2000" >H264 Mime Type = "video/H264" right, these are the RTP stream mime types, I think. > >Thanks and regards >-Lakshman > > > > > >--- Dave Singer wrote: > >> At 23:17 -0700 15/05/05, Alex wrote: >> >Hi All, >> > >> >is there any document specifying the syntax for >> video mime types (e. >> >g. H.263 / MPEG4, uppercase / lowercase, etc.) ? I >> was able to find >> >RFC 3555 - MIME Type Registration of RTP Payload >> Formats, but it >> >seems that's not actually what I'm looking for. >> > >> >> the MP4 file format registration is still, um, in >> process. It's >> supposed to be video/mp4, or for files with no >> visual aspect at all, >> audio/mp4 may be used also. >> >> >Thanks, >> > >> > >> > >> > >> >-- >> >Regards, >> >Alex >> > >> > >> > >> >Discover Yahoo! >> >Have fun online with music videos, cool games, IM & >> more. >> >>Check >> >> >it out! >> > >> >_______________________________________________ >> >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> >_______________________________________________ >> 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 > > > > >__________________________________ >Do you Yahoo!? >Yahoo! Small Business - Try our new resources site! >http://smallbusiness.yahoo.com/resources/ >_______________________________________________ >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 jinsp mcubeworks.com Tue May 17 23:04:16 2005 From: jinsp mcubeworks.com (Park Jin Soo) Date: Tue May 17 09:32:03 2005 Subject: [Mp4-tech] [Audio][BSAC] mismatch between frameLength and the decoded bytes Message-ID: <003501c55ae0$ee3bfc90$a900010a@mcubeworks.com> Dear experts, This is a question on mismatch between frameLength (the first member in bsac_base_element() or sam_decode_frame()) and the actually decoded bytes in sam_decode_frame and sam_decode_bsac_stream(). I tested if frameLength represents the length of a frame including bsac_header, general_header, bsac_layer_element (I summed all cw_len in sam_decode_frame()), byte_alignment, and frame_length itself with RefSoft and the conformance bitstreams. But the result shows that frameLength is larger than the actually decoded bytes for a few frames by 2 bytes at maximum. In other words, frameLength*8 is larger than the decoded bits for some frames by more than 7 bits. (16 bits at maximum). Here my question is: 1. What is wrong with it? - due to incorrect test or my mis-understanding of syntax? - or is it allowed that frameLength doen't represent the actual frame length? 2. What does frameLength mean for SBA? (For SBA, er_bs06*, frameLength has a quite large value as if it seems to do something for "the segments of a frame" rather than one frame itself) Thanks in advance, J.S Park From loloo 21cn.com Wed May 18 11:08:22 2005 From: loloo 21cn.com (loloo) Date: Wed May 18 05:35:06 2005 Subject: [Mp4-tech] Is it a bug in cavlc? Message-ID: <004501c55b4e$7b08fb10$512f60ac@EAT> In the ITU-T protocol H.264, Chapter 9.2.2 Parsing process for level information, It is said: When level_prefix is greater than or equal to 16, levelCode is incremented by (1<<( level_prefix �C 3 )) �C 4096. Let see when level_prefix is 16,17,18,19 and the value (1<<( level_prefix �C 3 )) �C 4096 is: (1)b<<12; (11)b<<12; (111)b<<12; (1111)b<<12; So levelCode is impossible for some value such as (10)b<<12, (100)b<<12 and so on. Let assume levelCode is (10,0000,0000,0000)b: If it is reduced by 1<<12( when level_prefix is 16),the result value is 1<<12 and is still more than 12 bits; If it is reduced by (11)b<<12( when level_prefix is 17),the result value is negative. So it is impossible to find a level_prefix to reduced it Is it a bug for encoder some large levelCode in cavlc? -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050518/ee8f9808/attachment.html From loloo 21cn.com Wed May 18 11:35:24 2005 From: loloo 21cn.com (loloo) Date: Wed May 18 05:35:11 2005 Subject: [Mp4-tech] Re: Is it a bug in cavlc? Message-ID: <005401c55b52$3ff93630$512f60ac@EAT> Sorry,I am wrong. The protocol is right,because in the new released protocol version it is: --When level_prefix is greater than or equal to 15, levelSuffixSize is set equal to level_prefix - 3; so it is not always equal to length 12 as the old version protocol is. ruilu -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050518/c8869530/attachment.html From pooya.dehghani gmail.com Wed May 18 01:49:21 2005 From: pooya.dehghani gmail.com (Pooya Dehghani) Date: Wed May 18 05:35:17 2005 Subject: [Mp4-tech] MPEG4 Encoder n Decoder source codes In-Reply-To: <20050517081407.29778.qmail@webmail30.rediffmail.com> References: <20050517081407.29778.qmail@webmail30.rediffmail.com> Message-ID: Dear Snajay Here you can download mpeg-4 codecs. Good Luck Pooya On 17 May 2024 08:14:07 -0000, guntupalli mrine sanjay wrote: > > > > Hi, > > So far i seen so many links. But i am not able to know which is use ful for > me. Any links or codes, please help me out. > > Thanks a lot, > Sanjay > > > _______________________________________________ > 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 pooya.dehghani gmail.com Thu May 19 01:41:59 2005 From: pooya.dehghani gmail.com (Pooya Dehghani) Date: Thu May 19 11:02:05 2005 Subject: [Mp4-tech] MPEG4 Encoder n Decoder source codes In-Reply-To: References: <20050517081407.29778.qmail@webmail30.rediffmail.com> Message-ID: Ooooooooooooops!!!!! I forgot to paste the URL address, so here it is : http://megaera.ee.nctu.edu.tw/mpeg/ thanks to mayank agarwal good luck every body On 5/18/05, Pooya Dehghani wrote: > Dear Snajay > > Here you can download mpeg-4 codecs. > > Good Luck > Pooya > > On 17 May 2024 08:14:07 -0000, guntupalli mrine sanjay > wrote: > > > > > > > > Hi, > > > > So far i seen so many links. But i am not able to know which is use ful for > > me. Any links or codes, please help me out. > > > > Thanks a lot, > > Sanjay > > > > > > _______________________________________________ > > 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 asic_soc 163.com Thu May 19 10:54:35 2005 From: asic_soc 163.com (Shen Sha) Date: Thu May 19 11:02:10 2005 Subject: [Mp4-tech] [video][mpeg2] Where can I download the mpeg2 TS stream? Message-ID: <003201c55c15$b9ec0ad0$453210ac@TMT.COM> Hi all experts, Could you tell me where can I find the mpeg2 TS stream and the reference software of TS demuxing ? Best Regard, Shen Sha -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050519/0437c897/attachment.html From amni_gupta yahoo.co.in Thu May 19 08:02:04 2005 From: amni_gupta yahoo.co.in (Amni Gupta) Date: Thu May 19 11:02:16 2005 Subject: [Mp4-tech] MP4 - Enc and Dec Message-ID: <20050519060204.37228.qmail@web8403.mail.in.yahoo.com> Hi; I want to develop MPEG-4 Video only natural SP Encoder and Decoder. I want to know that which are the modules common between them. For Example, how much portion of VLC encoding / decoding code can be reutilized among them. Regards Amni. Yahoo! India Matrimony: Find your life partneronline. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050519/a4d03708/attachment.html From amni_gupta yahoo.co.in Thu May 19 14:08:00 2005 From: amni_gupta yahoo.co.in (Amni Gupta) Date: Thu May 19 11:02:21 2005 Subject: [Mp4-tech] Small Doubt Message-ID: <20050519120800.18502.qmail@web8404.mail.in.yahoo.com> Hi; Please clarify that in MPEG-4 Video Simple Profile, the feature of Error Resilience is for only Decoder and the feature of Rate Control is for only Encoder. It is small and stupid doubt, but I want to clarify. Please respond. Thanks and Regards Amni. Yahoo! India Matrimony: Find your life partneronline. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050519/f26b7b1e/attachment.html From amni_gupta yahoo.co.in Thu May 19 14:22:46 2005 From: amni_gupta yahoo.co.in (Amni Gupta) Date: Thu May 19 11:02:26 2005 Subject: [Mp4-tech] Doubt- Please Write Message-ID: <20050519122246.51501.qmail@web8408.mail.in.yahoo.com> Hi; As I think in MPEG-4 Video Simple Profile, the Error Resilience feature is for Decoder and Rate Control is for Encoder. I mean that the implementation of Error Resilience feature is for only Decoder. It is not going to add any extra effort for Encoder part. In the same way, the implementation of Rate Control feature is for only Encoder. It is not going to add any extra effort for Decoder part. Please Clarify. Regards Amni Yahoo! India Matrimony: Find your life partneronline. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050519/005c6ba7/attachment.html From garysull windows.microsoft.com Thu May 19 19:18:27 2005 From: garysull windows.microsoft.com (Gary Sullivan) Date: Fri May 20 01:12:34 2005 Subject: [Mp4-tech] Small Doubt Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2460EA1F2EC@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> I'm not sure whether you are talking about specific features of the reference software, or more general conceptual features supported by the video coding standard. Error resilience, in general, is a function of both the encoder and the decoder (and the systems and channel between them). Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Amni Gupta Sent: Thursday, May 19, 2024 5:08 AM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] Small Doubt Hi; Please clarify that in MPEG-4 Video Simple Profile, the feature of Error Resilience is for only Decoder and the feature of Rate Control is for only Encoder. It is small and stupid doubt, but I want to clarify. Please respond. Thanks and Regards Amni. Yahoo! India Matrimony : Find your life partner online . -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050519/3897f914/attachment.html From bharatsoni gmail.com Fri May 20 10:19:28 2005 From: bharatsoni gmail.com (Bharat Soni) Date: Fri May 20 01:15:51 2005 Subject: [Mp4-tech] MP4 - Enc and Dec In-Reply-To: <20050519060204.37228.qmail@web8403.mail.in.yahoo.com> References: <20050519060204.37228.qmail@web8403.mail.in.yahoo.com> Message-ID: Hi, Encoder also perform decoding of the recently encoded frame. Hence I think at least following activities are common: * Inverse Quantization * IDCT * Motion Compensation Regards, Bharat On 5/19/05, Amni Gupta wrote: > Hi; > I want to develop MPEG-4 Video only natural SP Encoder and Decoder. I want > to know that which are the modules common between them. For Example, how > much portion of VLC encoding / decoding code can be reutilized among them. > > > Regards > Amni. > > Yahoo! India Matrimony: Find your life partner online. > _______________________________________________ > 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 bharatsoni gmail.com Fri May 20 10:31:54 2005 From: bharatsoni gmail.com (Bharat Soni) Date: Fri May 20 01:16:57 2005 Subject: [Mp4-tech] Small Doubt In-Reply-To: <20050519120800.18502.qmail@web8404.mail.in.yahoo.com> References: <20050519120800.18502.qmail@web8404.mail.in.yahoo.com> Message-ID: Hi, Error Resilience feature helps decoder to reconstruct the picture in case some data is lost. But the support for Error Resilience is provided by the Encoder only. Decoder can support/use Error Resilience only if it is provided by the encoder. Regarding Rate Control, it is the tool that is present on the encoder side only. It is used to control the encoding bitrate. Regards, Bharat On 5/19/05, Amni Gupta wrote: > Hi; > Please clarify that in MPEG-4 Video Simple Profile, the feature of Error > Resilience is for only Decoder and the feature of Rate Control is for only > Encoder. It is small and stupid doubt, but I want to clarify. Please > respond. > > Thanks and Regards > Amni. > > Yahoo! India Matrimony: Find your life partner online. > _______________________________________________ > 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 daljit.dhillon patni.com Fri May 20 11:37:58 2005 From: daljit.dhillon patni.com (Daljit Singh Dhillon) Date: Fri May 20 01:51:54 2005 Subject: FW: [Mp4-tech] Doubt- Please Write Message-ID: <200505200512.j4K5C95s015515@spzha1.patni.com> _____ From: Daljit Singh Dhillon [mailto:daljit.dhillon@patni.com] Sent: Friday, May 20, 2024 10:37 AM To: 'Amni Gupta' Subject: RE: [Mp4-tech] Doubt- Please Write Hi! As far as rate control is concerned. Yes.. it is Encoder's job. Decoder just have to have input buffer of size given by MPEG specification. As far as error resilience is concerned, it is the be taken care at both ends. Encoder has to ensure that it encodes each data partition independently. Only then decoder can decode them properly even in the case of error. Besides since RVLC's can to be used, it is for encoder to encode using of RVLC. In short Error resilience requires extra efforts for Encoder as well as decoder. Regards, -Daljit _____ From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Amni Gupta Sent: Thursday, May 19, 2024 5:53 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] Doubt- Please Write Hi; As I think in MPEG-4 Video Simple Profile, the Error Resilience feature is for Decoder and Rate Control is for Encoder. I mean that the implementation of Error Resilience feature is for only Decoder. It is not going to add any extra effort for Encoder part. In the same way, the implementation of Rate Control feature is for only Encoder. It is not going to add any extra effort for Decoder part. Please Clarify. Regards Amni Yahoo! India Matrimony: Find your life partner online . http://www.patni.com World-Wide Partnerships. World-Class Solutions. _____________________________________________________________________ This e-mail message may contain proprietary, confidential or legally privileged information for the sole use of the person or entity to whom this message was originally addressed. Any review, e-transmission dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you have received this e-mail in error kindly delete this e-mail from your records. If it appears that this mail has been forwarded to you without proper authority, please notify us immediately at netadmin@patni.com and delete this mail. _____________________________________________________________________ http://www.patni.com World-Wide Partnerships. World-Class Solutions. _____________________________________________________________________ This e-mail message may contain proprietary, confidential or legally privileged information for the sole use of the person or entity to whom this message was originally addressed. Any review, e-transmission dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you have received this e-mail in error kindly delete this e-mail from your records. If it appears that this mail has been forwarded to you without proper authority, please notify us immediately at netadmin@patni.com and delete this mail. _____________________________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050520/57b0b435/attachment.html From sandipkr dgbmicro.com Fri May 20 12:01:54 2005 From: sandipkr dgbmicro.com (Sandip Ray) Date: Fri May 20 01:55:34 2005 Subject: [video] RE: [Mp4-tech] Doubt- Please Write Message-ID: The Rate Control is required in the Encoder(not in Decoder) to generate bitstream of particular bitrate. Error Resilience feature is common to both Encoder and Decoder. For example, MPEG 4 supports Error Resilience features like Video Packet Resync Marker, Data Partitioning, Reversible Variable Length Code etc. The Error Concealment is the feature to be supported in Decoder only. The Error Concealment algorithms are not defined in the standard and any proprietary algorithm can be used. Regards, Sandip -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050520/12f1b9a4/attachment-0001.html From zhenzhongc hotmail.com Fri May 20 10:47:30 2005 From: zhenzhongc hotmail.com (Chen Zhenzhong) Date: Fri May 20 19:44:28 2005 Subject: [video][Mp4-tech] Help needed: MPEG-4 standard gray-level shape video sequence In-Reply-To: Message-ID: Dear All, I am trying to get the MPEG-4 standard gray-level shape video sequences (NOT the binary shape one). Can anybody here provide any hint? Thanks a lot. yours faithfully, Zhenzhong CHEN From gbmallikarjunarao yahoo.com Fri May 20 18:27:27 2005 From: gbmallikarjunarao yahoo.com (mallikarjun rao) Date: Fri May 20 19:44:33 2005 Subject: [Mp4-tech] Mp4-tech] [audio][MPEG4] AAC-LC,HE- AAC Decoder Message-ID: <20050520162727.49575.qmail@web52005.mail.yahoo.com> Hi, Can any tell, how to get AAC-LC,HE- AAC files of different bitrates? is there any tool to make AAC-LC,HE- AAC files with different bitrates.? Please reply asap. Thanks Mallikarjun --------------------------------- How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos. Get Yahoo! Photos -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050520/11c38603/attachment.html From wjianb buaa.edu.cn Mon May 23 07:22:23 2005 From: wjianb buaa.edu.cn (wjianb@buaa.edu.cn) Date: Sun May 22 20:30:14 2005 Subject: [Mp4-tech] the fps of FGS encoder Message-ID: <20050522222229.5355437F5A@mx2.buaa.edu.cn> hi,all experts, i tried to encode a test sequence "arkiyo_qcif" by a FGS-encoder which comes from Microsoft reference software (microsoft-v2.4-030710-NTU). It can work normally, but its fps which was diplayed on screen just be 1.23 ?? (and the decoder shows the fps is 2.37) why the fps value of the FGS encoder is so low? Best regards, wjianb wjianb@buaa.edu.cn 2024-05-23 From tom_mmedia yahoo.com Mon May 23 02:45:02 2005 From: tom_mmedia yahoo.com (Tomasz Dziecielewski) Date: Mon May 23 10:44:24 2005 Subject: [Mp4-tech] Problem with colocated blocks while encoding MBAFF frames Message-ID: <20050523084502.69507.qmail@web60923.mail.yahoo.com> Hello, I'm having difficult times trying to implement the direct prediction mode for MBAFF frames in an AVC encoder. After running some tests with the reference decoder, I'm almost sure that it is due to unproper determination of the colocated blocks, perhaps wrong colocated picture selection. My question thus concerns Table 8-6 in the AVC standard. I'm encoding a sequence composed of MBAFF frames only. Therefore, I assume that I should always pick the "The first entry in RefPicList1 is... A DECODED FRAME" and "field_pic_flag 0" entry in Table 8-6, choosing the firstRefPic1 (frame) as the colocated picture. But it appears that in this scenario, the reference decoder conducts the topAbsDiffPOC vs bottomAbsDiffPOC (both are always equal in our streams) comparison and picks one of the individual fields as the colocated picture. Or at least I'm guessing it does under all the mv table mangling that's going on in the code (inside the compute_colocated function). I'll be very grateful for any hints. Best regards, Tomasz Dziecielewski __________________________________ Do you Yahoo!? Make Yahoo! your home page http://www.yahoo.com/r/hs From mdc gatv.ssr.upm.es Mon May 23 14:19:55 2005 From: mdc gatv.ssr.upm.es (=?ISO-8859-1?Q?Mart=EDn_D=EDaz_Cuesta?=) Date: Mon May 23 10:44:30 2005 Subject: [Mp4-tech] The output for complete NALs Message-ID: <4291BC5B.2090206@gatv.ssr.upm.es> Hello experts! I would be pleased if you let me know the answer to this question: I am trying to separate the NAL flow before it is gathered in the file test.264. I know I must inspect the annexb.c file, and so I've done, but I can't still find where exactly (the concrete point) where the complete NAL is put into the file, and so where I should pick it up. Can you help me? Thanks on advance. Best regards for all From abdessamad.falhi epfl.ch Mon May 23 18:44:13 2005 From: abdessamad.falhi epfl.ch (abdessamad.falhi@epfl.ch) Date: Tue May 24 14:17:25 2005 Subject: [Mp4-tech] AAC decoder : raw_data_blocks format Message-ID: <1116863053.4291fa4d1380f@imapwww.epfl.ch> Hi everybody, Can anyone give me the raw_data_blocks format of an adts file? with regards, Abdessamad From alexismt comcast.net Mon May 23 10:04:34 2005 From: alexismt comcast.net (Alexis Michael Tourapis) Date: Tue May 24 14:21:07 2005 Subject: [Mp4-tech] Problem with colocated blocks while encoding MBAFF frames In-Reply-To: <20050523084502.69507.qmail@web60923.mail.yahoo.com> Message-ID: <200505231604.j4NG4eZ5006963@lists1.magma.ca> Dear Tomasz, When you mention MBAFF frames, are you implying that all Macroblocks are coded in frame mode as well? If that is not the case (most likely), then you need to also consider whether the co-located partition was in field or frame mode as well, and therefore also use the additional rules in table 8-6. Those rules were not specified only for picture level relationships, but also for MB level relationships as well. Best regards, Alexis -----Original Message----- From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Tomasz Dziecielewski Sent: Monday, May 23, 2024 1:45 AM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] Problem with colocated blocks while encoding MBAFF frames Hello, I'm having difficult times trying to implement the direct prediction mode for MBAFF frames in an AVC encoder. After running some tests with the reference decoder, I'm almost sure that it is due to unproper determination of the colocated blocks, perhaps wrong colocated picture selection. My question thus concerns Table 8-6 in the AVC standard. I'm encoding a sequence composed of MBAFF frames only. Therefore, I assume that I should always pick the "The first entry in RefPicList1 is... A DECODED FRAME" and "field_pic_flag 0" entry in Table 8-6, choosing the firstRefPic1 (frame) as the colocated picture. But it appears that in this scenario, the reference decoder conducts the topAbsDiffPOC vs bottomAbsDiffPOC (both are always equal in our streams) comparison and picks one of the individual fields as the colocated picture. Or at least I'm guessing it does under all the mv table mangling that's going on in the code (inside the compute_colocated function). I'll be very grateful for any hints. Best regards, Tomasz Dziecielewski __________________________________ Do you Yahoo!? Make Yahoo! your home page http://www.yahoo.com/r/hs _______________________________________________ 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 lucyamarnath yahoo.com Mon May 23 11:46:18 2005 From: lucyamarnath yahoo.com (amarnath nathan) Date: Tue May 24 14:23:03 2005 Subject: [Mp4-tech] regarding YUV Message-ID: <20050523174618.59921.qmail@unknown-209-73-178-238.yahoo.com> hello experts would u please tell me how to find the format of a yuv file by looking the file. i.e how to know whether it is 4:4:4 or 4:2:2 or 4:2:0.since i have a file in a format (qcif_15fps.yuv). Also plz suggest some ideas to convert .YUV to .SEQ. regards amar --------------------------------- Discover Yahoo! Have fun online with music videos, cool games, IM & more. Check it out! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050523/054d7d65/attachment.html From christophe.lenaerts telemak.com Tue May 24 00:39:09 2005 From: christophe.lenaerts telemak.com (Christophe Lenaerts) Date: Tue May 24 14:26:23 2005 Subject: [Mp4-tech] [video] mime types In-Reply-To: References: <20050516061725.74461.qmail@web53910.mail.yahoo.com> Message-ID: <92A5E064-815B-4A40-9AE2-19960E51983D@telemak.com> On 16 May 2005, at 21:46, Dave Singer wrote: >> >> is there any document specifying the syntax for video mime types >> (e. g. H.263 / MPEG4, uppercase / lowercase, etc.) ? I was able to >> find RFC 3555 - MIME Type Registration of RTP Payload Formats, but >> it seems that's not actually what I'm looking for. >> > > the MP4 file format registration is still, um, in process. It's > supposed to be video/mp4, or for files with no visual aspect at > all, audio/mp4 may be used also. What about 3GPP live streams with .sdp mime types? And .3gp VOD? Christophe ------------------------------------------------------------------------ --------------------- Christophe Lenaerts CEO - Telemak Support 0903 36 338 Fax (+32) 53 68 33 68 Mobile (+32) 475 47 18 15 ------------------------------------------------------------------------ --------------------- Sign in on our newsletter and keep up to date on what is going on at Telemak: telemaknews-on@lists.telemak.com ------------------------------------------------------------------------ --------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050523/c517926a/attachment.html From singer apple.com Mon May 23 16:02:48 2005 From: singer apple.com (Dave Singer) Date: Tue May 24 14:28:40 2005 Subject: [Mp4-tech] [video] mime types In-Reply-To: <92A5E064-815B-4A40-9AE2-19960E51983D@telemak.com> References: <20050516061725.74461.qmail@web53910.mail.yahoo.com> <92A5E064-815B-4A40-9AE2-19960E51983D@telemak.com> Message-ID: At 23:39 +0200 23/05/05, Christophe Lenaerts wrote: >On 16 May 2005, at 21:46, Dave Singer wrote: > >>> >>> >>>is there any document specifying the syntax for video mime types >>>(e. g. H.263 / MPEG4, uppercase / lowercase, etc.) ? I was able to >>>find RFC 3555 - MIME Type Registration of RTP Payload Formats, but >>>it seems that's not actually what I'm looking for. >>> >>> >>> >> >>the MP4 file format registration is still, um, in process. It's >>supposed to be video/mp4, or for files with no visual aspect at >>all, audio/mp4 may be used also. >> > >What about 3GPP live streams with .sdp mime types? And .3gp VOD? I don't understand the question. The mime type for .3gp files is video/3gpp. For streams, it depends on the codec what the type for the RTP stream is, and that's independent of whether it's 3G or MPEG-4 etc. > > >Christophe >--------------------------------------------------------------------------------------------- >Christophe Lenaerts >CEO - Telemak >Support 0903 36 338 >Fax (+32) 53 68 33 68 >Mobile (+32) 475 47 18 15 ><http://www.telemak.com> ><christophe.lenaerts@telemak.com> >--------------------------------------------------------------------------------------------- >Sign in on our newsletter and keep up to date on what is going on at >Telemak: >telemaknews-on@lists.telemak.com >--------------------------------------------------------------------------------------------- -- David Singer Apple Computer/QuickTime -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050523/a325732b/attachment.html From v_v_bhat yahoo.com Mon May 23 22:18:03 2005 From: v_v_bhat yahoo.com (vinayak bhat) Date: Tue May 24 14:30:41 2005 Subject: [Mp4-tech] Regarding H.263+(Annex K-Rectangular Slice Mode) Message-ID: <20050524041804.24333.qmail@web33010.mail.mud.yahoo.com> Hi All, I am facing problem in H.263 Annex K ( Rectangluar Slice Mode)..in that when I encode the first rectangle I am getting some blockage. & for the remaining rectangle's its not happening,...even the PSNR is going down by 1dB. thanks --vinayak __________________________________________________ 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/20050523/d2116af7/attachment.html From Xueyao9 netscape.net Tue May 24 05:56:14 2005 From: Xueyao9 netscape.net (Xueyao9@netscape.net) Date: Tue May 24 14:30:46 2005 Subject: [Mp4-tech] MPEG4 Audio SBR Conformance Tool Message-ID: <1171A339.5A5B1F2F.001BCCD8@netscape.net> Hello, Does anybody know how to compile the mpeg4 sbr conformance tool? I got quite a few compiling errors, such as: src/mp4dec.h:40: MP4Movies.h: No such file or directory src/audiofile_io.c:34: libtsp.h: No such file or directory src/audiofile_io.c:35: libtsp/AFpar.h: No such file or directory Pls help! And thanks a lot in advance! Robredo __________________________________________________________________ Switch to Netscape Internet Service. As low as $9.95 a month -- Sign up today at http://isp.netscape.com/register Netscape. Just the Net You Need. New! Netscape Toolbar for Internet Explorer Search from anywhere on the Web and block those annoying pop-ups. Download now at http://channels.netscape.com/ns/search/install.jsp From tom_mmedia yahoo.com Tue May 24 03:08:28 2005 From: tom_mmedia yahoo.com (Tomasz Dziecielewski) Date: Tue May 24 14:30:52 2005 Subject: [Mp4-tech] Problem with colocated blocks while encoding MBAFF frames Message-ID: <20050524090828.1169.qmail@web60921.mail.yahoo.com> Dear Alexis, Can you specify which of the rules in Table 8-6 are macroblock-wise? The way I see it (for the presented scenario) is: - field_pic_flag is always 0 for each coded picture is an MBAFF frame - the first entry in RefPicList1 is always a decoded frame for the same reason as above (I reckon an MBAFF frame can't be thought of as a complementary field pair?) Therefore the result is only picture-level dependent and the colocated picture is always the first frame in list 1, no matter whether the macroblock pair under encoding is a frame or field macroblock pair. Please correct me if I'm wrong. Best regards, Tomasz > When you mention MBAFF frames, are you implying that > all Macroblocks are > coded in frame mode as well? If that is not the case > (most likely), then you > need to also consider whether the co-located > partition was in field or frame > mode as well, and therefore also use the additional > rules in table 8-6. > Those rules were not specified only for picture > level relationships, but > also for MB level relationships as well. > > Best regards, > > Alexis __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new Resources site http://smallbusiness.yahoo.com/resources/ From tom_mmedia yahoo.com Tue May 24 03:27:19 2005 From: tom_mmedia yahoo.com (Tomasz Dziecielewski) Date: Tue May 24 14:30:58 2005 Subject: [Mp4-tech] Problem with colocated blocks while encoding MBAFF frames In-Reply-To: <42920398.4030700@polynet.lviv.ua> Message-ID: <20050524092719.27049.qmail@web60920.mail.yahoo.com> Dear Andrew, Yes, I'm aware of the constant assignment in the mbAddrCol6 derivation. As for the macroblock addressing itself, I assume it is always done with respect to the entire frame rather than its individual fields (for an MBAFF frame only sequence)? I'm asking this cause what bothers me is that the differences in the stream interpretation between our and the reference software occur for field macroblocks that have their colocated macroblocks being also field mbs, which appears to be a rather simple one-to-one mapping case. Is there some tricky part I'm not aware of? Or in other words, for the AFRM-AFRM case in Table 8-8, is the yM coordinate always given in the frame pixel units? Best regards, Tomasz Dziecielewski --- Andrew Voznytsa wrote: > Hello Tomasz, > > take a look on next case: > * > PicCodingStruct( CurrPic ) - *AFRM > *PicCodingStruct( colPic ) - ***AFRM > *mb_field_decoding_flag- 0 > **fieldDecodingFlagX - 1 > > In this case address of col mb is calculated > according to *mbAddrCol6 > formula: > > mbAddrCol6 = 2 * ( CurrMbAddr / 2 ) + ( ( > topAbsDiffPOC < > bottomAbsDiffPOC ) ? 0 : 1 ) > > In your case "topAbsDiffPOC < bottomAbsDiffPOC" is > always false so > bottom field should be used. > > Best regards, > Andrew Voznytsa __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new Resources site http://smallbusiness.yahoo.com/resources/ From sanjay_gims rediffmail.com Tue May 24 10:56:35 2005 From: sanjay_gims rediffmail.com (guntupalli mrine sanjay) Date: Tue May 24 14:31:06 2005 Subject: [Mp4-tech] Macroblock Message-ID: <20050524095813.14175.qmail@webmail8.rediffmail.com> ? Hi all, When i want to convert moving pictures to QCIF(176*144) and CIF (352*288)formats, Is MPEG4 saying that u need to convert that moving pictures to pictures and then pictures to macroblocks? Is the macroblock size is (8*8)? Let us say image is of QCIF(176*144)format. Then if i convert that to macroblocks of size 8*8 then some lines and pixels will be left over? How to handle those things? PLease give some hints how to convert Moving pictures to pictures? I am presently looking for MPEG4 Encoder and Decoder. Please give some information. Is there any opensources? Is it good to start from opensource? Let us say, if i want to develope MPEG4 Enco n Deco, how much time will it takes? PLease help me in this issue. Looking for response. Thanking you, G.I.M.sanjay. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050524/8bc430e5/attachment.html From m.semsar ece.ut.ac.ir Tue May 24 15:21:01 2005 From: m.semsar ece.ut.ac.ir (m.semsar@ece.ut.ac.ir) Date: Tue May 24 14:31:11 2005 Subject: [Mp4-tech] RDOptimization in Jm9.5 Message-ID: <24166.217.218.36.94.1116928261.squirrel@217.218.36.94> Dear all, Hi, I have switched form Jm9.3 to Jm9.5, so I have seen some new parameters added to encoder.cfg file, One of them is a new mode for rate distortion optimization which is explained as: # 2: RD-on (Fast high complexity mode - not work in FREX Profiles) What does this parameter mean? Thanks, M.Semsarzadeh From arjun cs.iitm.ernet.in Tue May 24 18:38:54 2005 From: arjun cs.iitm.ernet.in (T. Bheemarjuna Reddy) Date: Tue May 24 14:31:17 2005 Subject: [Mp4-tech] A doubt in calculation of PSNR value for Video In-Reply-To: <200505231606.j4NG1uZB006737@lists1.magma.ca> References: <200505231606.j4NG1uZB006737@lists1.magma.ca> Message-ID: Hi all, I have one basic doubt in calculating PSNR value. PSNR (db) = 10 log {(2^n - 1)^2 / MSE} what happens to value of PSNR if MSE = 0 ? I am encountering this scenario while transmiting video pkts over network. Here i am measuring PSNR of transmitted vs received video frames. Thanks inadvance. Regards Arjun Reddy -- ================================ = Bheemarjuna Reddy Tamma = = PhD Student = = HPCN Lab, CS&E Department = = IIT Madras, INDIA = = Phone: +91-44-22575369 (Lab) = = http://hpcn.cs.iitm.ernet.in = ================================ From abdessamad.falhi epfl.ch Tue May 24 16:19:12 2005 From: abdessamad.falhi epfl.ch (abdessamad.falhi@epfl.ch) Date: Tue May 24 14:31:24 2005 Subject: [Mp4-tech] AAC decoder : DRC Message-ID: <1116940752.429329d09de74@imapwww.epfl.ch> Hi, Does anyone know why DRC is used in aac decoder, and can we let it down whithout violating the aac normes? with regards, Abdessamad From mp3.aac.mp4 gmail.com Tue May 24 20:01:37 2005 From: mp3.aac.mp4 gmail.com (tech list) Date: Tue May 24 14:31:29 2005 Subject: [Mp4-tech] [audio][MPEG4]SBR HF generation and Adjustment In-Reply-To: <20050513052456.54051.qmail@web20126.mail.yahoo.com> References: <20050513052456.54051.qmail@web20126.mail.yahoo.com> Message-ID: <409a09b9050524063166c6f1c7@mail.gmail.com> On 5/13/05, Dattaguru B.N. wrote: > > Dear Experts, > > Can anybody please let me know, > 1. What informations of low frequency are required for > HF generation and adjustment? You need to refer to the 3GPP docs for enhanced aacplus for this... 2. What parameters during encoding decide the High > frequency generation and adjustment? As above 3. If I have an AAC bit stream, can I encode the same > using SBR encoder? In that case, what will be > difference between AAC bit stream compared to same bit > stream encoded with SBR? SBR is a post processing tool, so there is no seperate SBR encoder as such. AACplus or HE-AAC combines AAC and SBR. The SBR info is placed in fill elements in the AAC bitstream. 4. Whose audio quality is better between normal AAC > when compared with SBR and why? If they are the same bit rate, then obvisouly the SBR processed one will be better. Thanks in advance, > Warm regards, > Dattaguru > > > > Yahoo! Mail > Stay connected, organized, and protected. Take the tour: > http://tour.mail.yahoo.com/mailtour.html > > _______________________________________________ > 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/20050524/de1021e9/attachment.html From tariq ece.gatech.edu Tue May 24 15:33:13 2005 From: tariq ece.gatech.edu (Tarik Arici) Date: Wed May 25 06:20:58 2005 Subject: [Mp4-tech] regarding YUV In-Reply-To: <20050523174618.59921.qmail@unknown-209-73-178-238.yahoo.com> References: <20050523174618.59921.qmail@unknown-209-73-178-238.yahoo.com> Message-ID: <42937369.30505@ece.gatech.edu> you can read them in the 3 different formats and view the frames. whichever one looks normal will be the right format. Since there is no any header I dont think there is any other way amarnath nathan wrote: > hello experts > > would u please tell me how to find the format of a yuv file by looking > the file. i.e how to know whether it is 4:4:4 or 4:2:2 or 4:2:0.since > i have a file in a format (qcif_15fps.yuv). > > Also plz suggest some ideas to convert .YUV to .SEQ. > > regards > amar > > ------------------------------------------------------------------------ > Discover Yahoo! > Have fun online with music videos, cool games, IM & more. Check it > out! > > >------------------------------------------------------------------------ > >_______________________________________________ >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 garysull windows.microsoft.com Tue May 24 14:18:46 2005 From: garysull windows.microsoft.com (Gary Sullivan) Date: Wed May 25 06:26:57 2005 Subject: [Mp4-tech] Problem with colocated blocks while encoding MBAFFframes Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2460EB0695F@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> When decoding an MBAFF frame, the first picture in list 1 can be either a complementary reference field pair or an MBAFF reference frame. It is correct to say that an MBAFF frame cannot be thought of as a complementary field pair. They are two different things. (If they weren't, table 8-6 wouldn't have both cases in it.) I suggest making sure you have a recent (2005) draft. There were some changes in that table. Best regards, Gary Sullivan +> -----Original Message----- +> From: mp4-tech-bounces@lists.mpegif.org +> [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of +> Tomasz Dziecielewski +> Sent: Tuesday, May 24, 2024 2:08 AM +> To: mp4-tech@lists.mpegif.org +> Subject: RE: [Mp4-tech] Problem with colocated blocks while +> encoding MBAFFframes +> +> Dear Alexis, +> +> Can you specify which of the rules in Table 8-6 are +> macroblock-wise? The way I see it (for the presented +> scenario) is: +> - field_pic_flag is always 0 for each coded picture is +> an MBAFF frame +> - the first entry in RefPicList1 is always a decoded +> frame for the same reason as above (I reckon an MBAFF +> frame can't be thought of as a complementary field +> pair?) +> Therefore the result is only picture-level dependent +> and the colocated picture is always the first frame in +> list 1, no matter whether the macroblock pair under +> encoding is a frame or field macroblock pair. Please +> correct me if I'm wrong. +> +> Best regards, +> Tomasz +> +> > When you mention MBAFF frames, are you implying that +> > all Macroblocks are +> > coded in frame mode as well? If that is not the case +> > (most likely), then you +> > need to also consider whether the co-located +> > partition was in field or frame +> > mode as well, and therefore also use the additional +> > rules in table 8-6. +> > Those rules were not specified only for picture +> > level relationships, but +> > also for MB level relationships as well. +> > +> > Best regards, +> > +> > Alexis +> +> +> +> +> __________________________________ +> Do you Yahoo!? +> Yahoo! Small Business - Try our new Resources site +> http://smallbusiness.yahoo.com/resources/ +> _______________________________________________ +> 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 alexismt comcast.net Tue May 24 22:46:20 2005 From: alexismt comcast.net (Alexis Michael Tourapis) Date: Wed May 25 06:30:20 2005 Subject: [Mp4-tech] Problem with colocated blocks while encoding MBAFF frames Message-ID: <052420052146.25524.4293A0AC0002636C000063B422007347489B039C07980A040E@comcast.net> Dear Tomasz, I should have said table 8.8 not 8.6. Sorry for the typo. If you examine the very last row of this table you have an entry where both current and co-located are AFRM( as is your case). Different conditions apply depending on whether the current block is in field or frame mode and on the mode of the co-located. Note also that here the equations for mbAddrColX also apply which contain the relationships for top and bottom poc which are used by the software. I hope his helps, Alexis -------------- Original message -------------- > Dear Alexis, > > Can you specify which of the rules in Table 8-6 are > macroblock-wise? The way I see it (for the presented > scenario) is: > - field_pic_flag is always 0 for each coded picture is > an MBAFF frame > - the first entry in RefPicList1 is always a decoded > frame for the same reason as above (I reckon an MBAFF > frame can't be thought of as a complementary field > pair?) > Therefore the result is only picture-level dependent > and the colocated picture is always the first frame in > list 1, no matter whether the macroblock pair under > encoding is a frame or field macroblock pair. Please > correct me if I'm wrong. > > Best regards, > Tomasz > > > When you mention MBAFF frames, are you implying that > > all Macroblocks are > > coded in frame mode as well? If that is not the case > > (most likely), then you > > need to also consider whether the co-located > > partition was in field or frame > > mode as well, and therefore also use the additional > > rules in table 8-6. > > Those rules were not specified only for picture > > level relationships, but > > also for MB level relationships as well. > > > > Best regards, > > > > Alexis > > > > > __________________________________ > Do you Yahoo!? > Yahoo! Small Business - Try our new Resources site > http://smallbusiness.yahoo.com/resources/ > _______________________________________________ > 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/20050524/1742f62c/attachment-0001.html From alexismt comcast.net Tue May 24 22:59:10 2005 From: alexismt comcast.net (Alexis Michael Tourapis) Date: Wed May 25 06:30:27 2005 Subject: [Mp4-tech] Problem with colocated blocks while encoding MBAFF frames Message-ID: <052420052159.13291.4293A3AE00035BBE000033EB22007347489B039C07980A040E@comcast.net> Dear Tomasz, Although admitedly it took us quite some time to finish this part of the software(especially since this was done on "volunteering" terms), the software should currently be correctly following the spec. However it seems to me that maybe the reason why you might be having a problem is your interpretation of yM. If you carefully read section 6.4.3 you will notice that the (x,y) coordinates (and xCol, yCol, and yM), are relative to a Macroblock not the frame. Your email suggests otherwise. Alexis -------------- Original message -------------- > Dear Andrew, > > Yes, I'm aware of the constant assignment in the > mbAddrCol6 derivation. As for the macroblock > addressing itself, I assume it is always done with > respect to the entire frame rather than its individual > fields (for an MBAFF frame only sequence)? I'm asking > this cause what bothers me is that the differences in > the stream interpretation between our and the > reference software occur for field macroblocks that > have their colocated macroblocks being also field mbs, > which appears to be a rather simple one-to-one mapping > case. Is there some tricky part I'm not aware of? Or > in other words, for the AFRM-AFRM case in Table 8-8, > is the yM coordinate always given in the frame pixel > units? > > Best regards, > Tomasz Dziecielewski > > --- Andrew Voznytsa wrote: > > Hello Tomasz, > > > > take a look on next case: > > * > > PicCodingStruct( CurrPic ) - *AFRM > > *PicCodingStruct( colPic ) - ***AFRM > > *mb_field_decoding_flag- 0 > > **fieldDecodingFlagX - 1 > > > > In this case address of col mb is calculated > > according to *mbAddrCol6 > > formula: > > > > mbAddrCol6 = 2 * ( CurrMbAddr / 2 ) + ( ( > > topAbsDiffPOC < > > bottomAbsDiffPOC ) ? 0 : 1 ) > > > > In your case "topAbsDiffPOC < bottomAbsDiffPOC" is > > always false so > > bottom field should be used. > > > > Best regards, > > Andrew Voznytsa > > > > > __________________________________ > Do you Yahoo!? > Yahoo! Small Business - Try our new Resources site > http://smallbusiness.yahoo.com/resources/ > _______________________________________________ > 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/20050524/ba98372e/attachment.html From alexismt comcast.net Tue May 24 23:08:12 2005 From: alexismt comcast.net (Alexis Michael Tourapis) Date: Wed May 25 06:30:33 2005 Subject: [Mp4-tech] RDOptimization in Jm9.5 Message-ID: <052420052208.25043.4293A5CC0001F5B5000061D322007347489B039C07980A040E@comcast.net> RD mode 2 is part of the implementation of the Fast High Complexity mode as this is described in the document "Text Description of Joint Model Reference Encoding Methods and Decoding Concealment Methods". You can find this document here (see section 2.1.3) : http://ftp3.itu.ch/av-arch/jvt-site/2005_01_HongKong/JVT-N046r1.doc For reference, the algorithms described in these document are for reference only. It does not mean you have to follow these algorithms or that you cannot do considerably better if you use a different approach. Best regards, Alexis -------------- Original message -------------- > Dear all, > Hi, > I have switched form Jm9.3 to Jm9.5, so I have seen some new parameters > added to encoder.cfg file, > One of them is a new mode for rate distortion optimization which is > explained as: > # 2: RD-on (Fast high complexity mode - not work in FREX Profiles) > What does this parameter mean? > Thanks, > M.Semsarzadeh > > > _______________________________________________ > 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/20050524/d81d93a8/attachment.html From braindumper gmail.com Wed May 25 10:25:57 2005 From: braindumper gmail.com (Aj Bangla) Date: Wed May 25 06:30:39 2005 Subject: [Mp4-tech] Re: AAC decoder : raw_data_blocks format In-Reply-To: <200505242013.j4OKCh3P021397@lists1.magma.ca> References: <200505242013.j4OKCh3P021397@lists1.magma.ca> Message-ID: All the formats have same syntax for raw data block which is given in subpart 4 > Message: 1 > Date: Mon, 23 May 2024 17:44:13 +0200 > From: abdessamad.falhi@epfl.ch > Subject: [Mp4-tech] AAC decoder : raw_data_blocks format > To: mp4-tech@lists.mpegif.org > Message-ID: <1116863053.4291fa4d1380f@imapwww.epfl.ch> > Content-Type: text/plain; charset=ISO-8859-1 > > > > Hi everybody, > > Can anyone give me the raw_data_blocks format of an adts file? > > with regards, > > Abdessamad > > -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050525/50bfdad5/attachment.html From kannan.g.s.nambiar celstream.com Wed May 25 11:03:38 2005 From: kannan.g.s.nambiar celstream.com (Kannan GS Nambiar) Date: Wed May 25 06:30:46 2005 Subject: [Mp4-tech] MPEG4 Audio SBR Conformance Tool Message-ID: <80464F9A4D2BF042A154DD067F1F539311C4FF@cel-bangt-m01.india.celstream.com> Hi, I think you can try this link for LIBTSP package http://www.tsp.ece.mcgill.ca/MMSP/Documents/index.html and If I remember well MP4Movies.h is part of MPEG4 Systems. Regards, Kannan -----Original Message----- From: Xueyao9@netscape.net [mailto:Xueyao9@netscape.net] Sent: Tuesday, May 24, 2024 2:26 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] MPEG4 Audio SBR Conformance Tool Hello, Does anybody know how to compile the mpeg4 sbr conformance tool? I got quite a few compiling errors, such as: src/mp4dec.h:40: MP4Movies.h: No such file or directory src/audiofile_io.c:34: libtsp.h: No such file or directory src/audiofile_io.c:35: libtsp/AFpar.h: No such file or directory Pls help! And thanks a lot in advance! Robredo __________________________________________________________________ Switch to Netscape Internet Service. As low as $9.95 a month -- Sign up today at http://isp.netscape.com/register Netscape. Just the Net You Need. New! Netscape Toolbar for Internet Explorer Search from anywhere on the Web and block those annoying pop-ups. Download now at http://channels.netscape.com/ns/search/install.jsp _______________________________________________ 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 is free from Virus - IMSS -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050525/cbce45a6/attachment.html From snd codingtechnologies.com Wed May 25 10:14:03 2005 From: snd codingtechnologies.com (Andreas Schneider) Date: Wed May 25 06:30:51 2005 Subject: [Mp4-tech] MPEG4 Audio SBR Conformance Tool In-Reply-To: <1171A339.5A5B1F2F.001BCCD8@netscape.net> Message-ID: Robredo, the SBR conformance tool requires - the mp4 file-format library - the AFsp audio file libary Regards, Andreas mp4-tech-bounces@lists.mpegif.org wrote on 2024-05-24 10:56:14: > Hello, > > Does anybody know how to compile the mpeg4 sbr conformance tool? I > got quite a few compiling errors, such as: > > src/mp4dec.h:40: MP4Movies.h: No such file or directory > src/audiofile_io.c:34: libtsp.h: No such file or directory > src/audiofile_io.c:35: libtsp/AFpar.h: No such file or directory > > Pls help! And thanks a lot in advance! > Robredo > > __________________________________________________________________ > Switch to Netscape Internet Service. > As low as $9.95 a month -- Sign up today at http://isp.netscape.com/register > > Netscape. Just the Net You Need. > > New! Netscape Toolbar for Internet Explorer > Search from anywhere on the Web and block those annoying pop-ups. > Download now at http://channels.netscape.com/ns/search/install.jsp > _______________________________________________ > 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 tuukkat ee.oulu.fi Wed May 25 11:31:13 2005 From: tuukkat ee.oulu.fi (Tuukka Toivonen) Date: Wed May 25 06:30:58 2005 Subject: [Mp4-tech] A doubt in calculation of PSNR value for Video In-Reply-To: References: <200505231606.j4NG1uZB006737@lists1.magma.ca> Message-ID: On Tue, 24 May 2005, T. Bheemarjuna Reddy wrote: > I have one basic doubt in calculating PSNR value. > > PSNR (db) = 10 log {(2^n - 1)^2 / MSE} > > what happens to value of PSNR if MSE = 0 ? The PSNR is infinite then. You could check specifically for this case and return the largest possible number. If using IEEE floating points, they can represent infinite exactly. However, normally this is very rare case so people often ignore it. From rajeshnair itmagic.co.kr Wed May 25 17:59:57 2005 From: rajeshnair itmagic.co.kr (Rajesh G Nair) Date: Wed May 25 06:31:06 2005 Subject: [Mp4-tech] Extracting video from Divx file In-Reply-To: <4288E486.3040408@iis.fhg.de> Message-ID: <200505250752.j4P7q3fj023473@ns8.nic21c.com> Hi, I was wondering how I could parse the video information form the divx file, so that I could play the M4v file with my mpeg4 decoder It would be of great help if anyone could help me out on this Thanx in advance rajesh From dattagurubn yahoo.com Wed May 25 05:54:06 2005 From: dattagurubn yahoo.com (Dattaguru B.N.) Date: Wed May 25 08:12:27 2005 Subject: [Mp4-tech] [audio][MPEG4]SBR HF generation and Adjustment In-Reply-To: <409a09b9050524063166c6f1c7@mail.gmail.com> Message-ID: <20050525115406.41086.qmail@web20125.mail.yahoo.com> Dear ALL, Thanks. I have one more question. Is there any limitation for the Envelope Mapped value used in calculate gain of HF Adjustment. If limitation is there, what is the value? (note: I am getting the value of the range of 1e+12. Is this value possible?) THanks a lot for the quick response. Warm regards, Dattaguru --- tech list wrote: > On 5/13/05, Dattaguru B.N. > wrote: > > > > Dear Experts, > > > > Can anybody please let me know, > > 1. What informations of low frequency are required > for > > HF generation and adjustment? > > > You need to refer to the 3GPP docs for enhanced > aacplus for this... > > 2. What parameters during encoding decide the High > > frequency generation and adjustment? > > > As above > > 3. If I have an AAC bit stream, can I encode the > same > > using SBR encoder? In that case, what will be > > difference between AAC bit stream compared to same > bit > > stream encoded with SBR? > > > SBR is a post processing tool, so there is no > seperate SBR encoder as such. > AACplus or HE-AAC combines AAC and SBR. The SBR info > is placed in fill > elements in the AAC bitstream. > > 4. Whose audio quality is better between normal AAC > > when compared with SBR and why? > > > If they are the same bit rate, then obvisouly the > SBR processed one will be > better. > > Thanks in advance, > > Warm regards, > > Dattaguru > > > > > > > > Yahoo! Mail > > Stay connected, organized, and protected. Take the > tour: > > http://tour.mail.yahoo.com/mailtour.html > > > > _______________________________________________ > > 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 __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new Resources site http://smallbusiness.yahoo.com/resources/ From mihir ti.com Wed May 25 19:46:42 2005 From: mihir ti.com (Mody, Mihir) Date: Thu May 26 03:49:59 2005 Subject: [Mp4-tech] Extracting video from Divx file Message-ID: Hi, You can use DirectX filters to parse DIVx file. Regards, Mihir Mody, -----Original Message----- From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Rajesh G Nair Sent: Wednesday, May 25, 2024 1:30 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] Extracting video from Divx file Hi, I was wondering how I could parse the video information form the divx file, so that I could play the M4v file with my mpeg4 decoder It would be of great help if anyone could help me out on this Thanx in advance rajesh _______________________________________________ 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 sathishn teneoris.com Wed May 25 14:55:02 2005 From: sathishn teneoris.com (Sathish N) Date: Thu May 26 03:50:06 2005 Subject: [Mp4-tech] Conformance testing of MPEG4 Advanced Simple profile (Video) Message-ID: <1117029283.27518.285.camel@sathish> Hello, I'm looking for Conformance tests & bitstreams for MPEG-4 video for Advanced Simple profile. In this process, I purchased ISO Conformance document (ISO/IEC 14496-4) & the test-streams. 1. I found 13 bitstreams under the directory 'advanced_simple'. But the Conformance document does not have any description about these bitstreams. No Table given in the document has any mention on 'Advanced Simple profile'. 2. I came across a link : 'MPEG-4 Video Conformance Bitstreams available for download' in MPEG4 Industry forum (m4if.org). Under this link also, I found some bit-streams under the directory 'Advanced_Simple'. These streams are different from the ones which are distributed from ISO. Also, these appear to be newer than the ones from ISO (as evident from the timestamps of the files). 3. Eventhough ISO/IEC 14496-4 document does not seem to categorize any bitstreams under Advanced Simple profile, there are some bit-streams listed under Section 5.6 - 'Additional Conformance Testing' which have the features of Advanced Simple profile like Quarter-sample interpolation & GMC. In view of the above, I'm not quite clear about which of the sources I should be using (1, 2 or 3 mentioned above) for testing the features of my Advanced Simple profile decoder. Would some one please clarify ? Regards, Sathish From almomo1 gmail.com Wed May 25 17:45:59 2005 From: almomo1 gmail.com (Alejandro Moya) Date: Thu May 26 03:50:12 2005 Subject: [Mp4-tech][Video H.264] Error in one quarter interpolation for luma samples Message-ID: <84ac0130505250745512691d9@mail.gmail.com> Hi, everyone I have studying H.264 standard lately and I discovered yesterday what I guess it could be an error: when you apply the 6-tap filter to get the pixel labeled as "j" in Figure 8-4, the formulae is identical to the one you have to apply to get "b" or "h" with one huge difference: for obtaining the last ones you do b = Clip1( (b1 + 16)>>5 ) (8-187) h = Clip1( (h1 + 16)>>5 ) (8-188) and to obtain "j" you do: j = Clip1((j1 + 512)>>10) This means that in the first case the filter is modulo 1, but in the second is modulo 1/32. I think the proper formulae to get "j" should be: j = Clip1( (j1 + 16)>>5 ) (8-191) Let's say that full pel samples G, H, M and N (the surrounding ones to "j") have all the value 0xB2, so common sense tells us that the luma pixel at 1/4 position (the one called "j") should be 0xB2 too. But applying equation (8-191) we get the value ox 0x06 at that 1/4 interpolated position, so it makes no sense. I displayed the interpolated image and got some ugly "black dots" at that interpolated positions. I would like to ask all of you what do you think about that. Maybe I am missing some littled detail when interpolating "j" or maybe standard is wrong? I have the last revision which I just downloaded today from ITU-T webpage. Thanks you!! From mahdi essex.ac.uk Wed May 25 19:12:06 2005 From: mahdi essex.ac.uk (M Mahdi Ghandi) Date: Thu May 26 03:50:17 2005 Subject: [Mp4-tech] A doubt in calculation of PSNR value for Video In-Reply-To: Message-ID: <002a01c5614c$e0b82360$4c40f59b@essex.ac.uk> The main problem is when for example this happened for only one frame in a sequence. I will not include that infinite in average PSNR calculation or alternatively replace it with an assumed upper limit (e.g. 50 dB). -----Original Message----- From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Tuukka Toivonen Sent: 25 May 2024 08:31 To: T. Bheemarjuna Reddy Cc: mp4-tech@lists.mpegif.org Subject: Re: [Mp4-tech] A doubt in calculation of PSNR value for Video On Tue, 24 May 2005, T. Bheemarjuna Reddy wrote: > I have one basic doubt in calculating PSNR value. > > PSNR (db) = 10 log {(2^n - 1)^2 / MSE} > > what happens to value of PSNR if MSE = 0 ? The PSNR is infinite then. You could check specifically for this case and return the largest possible number. If using IEEE floating points, they can represent infinite exactly. However, normally this is very rare case so people often ignore it. _______________________________________________ NOTE: Please use clear subject lines for your posts. Include [audio, [video], [systems], [general] or another apppropriate identifier to indicate the type of question you have. Note: Conduct on the mailing list is subject to the Antitrust guidelines found at http://www.mpegif.org/public/documents/vault/mp-out-30042-Antitrust.php From tuukkat ee.oulu.fi Wed May 25 21:22:02 2005 From: tuukkat ee.oulu.fi (Tuukka Toivonen) Date: Thu May 26 03:50:22 2005 Subject: [Mp4-tech] A doubt in calculation of PSNR value for Video In-Reply-To: <002a01c5614c$e0b82360$4c40f59b@essex.ac.uk> References: <002a01c5614c$e0b82360$4c40f59b@essex.ac.uk> Message-ID: On Wed, 25 May 2005, M Mahdi Ghandi wrote: > The main problem is when for example this happened for only one frame in a > sequence. I will not include that infinite in average PSNR calculation or > alternatively replace it with an assumed upper limit (e.g. 50 dB). Obviously average PSNR is a flawed measure. One really should compute MSE over all of the frames, and only then compute the PSNR. However, as JM and, if I recall correctly, other reference encoders use average PSNR, we're pretty much stuck with it. However, in typical cases average PSNR has very little difference from properly computed PSNR. From S.T.C.Beesley lboro.ac.uk Wed May 25 20:02:28 2005 From: S.T.C.Beesley lboro.ac.uk (Steve) Date: Thu May 26 03:50:28 2005 Subject: [Mp4-tech] A doubt in calculation of PSNR value for Video In-Reply-To: Message-ID: <200505251802.j4PI2dMH029790@lists1.magma.ca> Not sure about the decoder but at the encoder stage the value of MSE is checked prior to the PSNR calculation. If the MSE is 0 then it is set to 1, this is done in image.c - find_snr(). This avoids any problems representing an infinite PSNR. HTH, Steve -----Original Message----- From: T. Bheemarjuna Reddy [mailto:arjun@cs.iitm.ernet.in] Sent: 24 May 2024 13:09 To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] A doubt in calculation of PSNR value for Video Hi all, I have one basic doubt in calculating PSNR value. PSNR (db) = 10 log {(2^n - 1)^2 / MSE} what happens to value of PSNR if MSE = 0 ? I am encountering this scenario while transmiting video pkts over network. Here i am measuring PSNR of transmitted vs received video frames. Thanks inadvance. Regards Arjun Reddy -- ================================ = Bheemarjuna Reddy Tamma = = PhD Student = = HPCN Lab, CS&E Department = = IIT Madras, INDIA = = Phone: +91-44-22575369 (Lab) = = http://hpcn.cs.iitm.ernet.in = ================================ From garysull windows.microsoft.com Wed May 25 14:55:29 2005 From: garysull windows.microsoft.com (Gary Sullivan) Date: Thu May 26 03:50:33 2005 Subject: [Mp4-tech] Introductory information about the JVT and its work (2024-05-25) Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2460EB5FECB@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> The following information about the Joint Video Team (JVT) and its work may be helpful to some of you. [version: 2024-05-25] The primary work of the JVT currently consists of: 1) scalable video coding (SVC) extension development, and 2) maintenance of the existing Advanced Video Coding (AVC) standard ITU-T Rec. H.264 & ISO/IEC 14496-10, e.g., including errata reporting and maintenance of reference sotware and conformance specifications. The JVT currently has 3 active email reflectors. You can subscribe to two of them (the general JVT reflector and the conformance/interop bitstream exchange activity reflector) through http://mailman.rwth-aachen.de/mailman/listinfo/jvt-experts and http://mailman.rwth-aachen.de/mailman/listinfo/jvt-bitstream To subscribe to the 3rd JVT reflector (which is devoted to SVC work), send email to "majordomo@ient.rwth-aachen.de" containing "subscribe svc" in the body of the message. JVT documents can be found at http://ftp3.itu.int/av-arch/jvt-site No password is required for access to nearly all documents on that site. A select few documents (such as integrated-format standard drafts) require password access, using a password given only to formal JVT members. That document archive can alternatively be accessed by ftp using the same top-level site address (ftp3.itu.int) with the user ID "avguest" and the password "Avguest". This accesses the file system at the same level as the http access point http://ftp3.itu.int/av-arch, so the directory "jvt-site" there contains the JVT document archive (the other directories there are are for various ITU-T groups -- note that "video-site" there is for ITU-T VCEG, not the JVT). That user ID and password can be considered PUBLIC information. The JVT reference software coordination site for non-SVC work is: http://iphome.hhi.de/suehring/tml/ No password is required for access to that site. That software site does not have ftp access capability. The JVT reference software access information for SVC work is: Document JVT-N024 contains the JSVM 1 software. The previous major version was called JSVM 0, and is found in document JVT-N022. Those "documents" are found in http://ftp3.itu.ch/av-arch/jvt-site/2005_01_HongKong A new major release should be arriving soon. A CVS server for the JSVM software has been set up at RWTH Aachen, hosted at garcon.ient.rwth-aachen.de. It can be accessed using WinCVS or any other CVS client using the "pserver" method, with the user ID "jvtuser" and the password "jvt.Amd.2". The path is "/cvs/jvt", and the module name is "jsvm". For example, with a command line client, you can check out the software using the following commands. "cvs -d :pserver:jvtuser@garcon.ient.rwth-aachen.de:/cvs/jvt login" and "cvs -d :pserver:jvtuser@garcon.ient.rwth-aachen.de:/cvs/jvt co jsvm" (you can use "co" or "checkout") The server is configured to allow read access using the above-described jvtuser user ID. Write access to the SVC software server is restricted to JVT software coordinators Justin Ridge, Heiko Schwarz, and Mathias Wien. The next JVT meeting will be in Poznan Poland. The dates that were preliminarily announced for that meeting were 23-29 July. However, there is a substantial likelihood that we will postpone the start of the meeting until the 24th. The document registration and upload date is likely to be Tuesday July 19 or Wednesday July 20 at midnight Poznan time. After the Poznan meeting, the plan for the next two JVT meetings will be to co-locate them with each MPEG meeting (i.e., 16-21 October in Nice France and 15-20 January in Bangkok Thailand). That co-location of meetings is expected to continue until the 1st 2006 meeting of ITU-T SG 16, upon which we plan to meet alongside SG16, approximately 12-17 March 2006. We are then likely to return to meeting with MPEG (16-21 July in Klagenfurt Austria and 22-27 October in Hangzhou China). The JVT has two parent bodies, which are MPEG (ISO/IEC JTC 1/SC 29/WG 11) and VCEG (ITU-T SG 16 Q.6). Participation in the JVT is open to anyone who is qualfied to participate either in MPEG or VCEG, and to those personally invited by the chairmen. We are liberal in granting invitation requests. To progress the work of the JVT between meetings, the JVT has created the following ad-hoc groups, and has appointed the following listed chairpersons for that work. The discussions involved in the work of those ad-hoc groups will be conducted on the above-listed email reflectors. 1. JVT Project Management and Errata Reporting (Gary Sullivan, Jens Rainer Ohm, Ajay Luthra, and Thomas Wiegand) 2. JM Description and Reference Software (Thomas Wiegand, Karsten S?hring, Alexis Tourapis, and Keng Pang Lim) 3. Bitstream Exchange and Conformance (Teruhiko Suzuki and Lowell Winger) 4. SVC Core Experiments (Justin Ridge, Ulrich Benzler) 5. JSVM software improvement and new functionality integration (Greg Cook) 6. JSVM Text and WD Text Editing (Julien Reichel, Heiko Schwarz, Mathias Wien) 7. Spatial Scalability Resampling Filters (Gary Sullivan) 8. Test conditions and applications for error resilience (Ye Kui Wang) 9. Test conditions for coding efficiency work and JSVM performance evaluation (Mathias Wien, Heiko Schwarz) 10. Study of 4:4:4 video coding functionality (Teruhiko Suzuki) In the work on scalable video coding (SVC), the JVT is conducting the following core experiments (CEs). A document describing each of these CEs is available on the JVT ftp site in the 2005_04_Busan directory as document number JVT-O3xx, where "xx" is the number of the core experiment as listed below. The appointed core experiment coordinator, some participating companies, and some relevant documents (prefix the numbers below by "JVT-O" for the complete document number) are also listed below. CE1: MCTF memory management (009, 026, 027, 028) (Visiowave, Panasonic, Nokia) Julien Reichel CE2: Improved de-blocking filter settings (non-normative?) (RWTH, FTRD) (067) Mathias Wien CE3: Coding efficiency of entropy coding (SKKU, ETRI, Samsung) Woong Il Choi, (021, 063) CE4: Inter-layer motion prediction (Samsung, LG) Kyohyuk Lee (058) CE5: Quality Layers (FTRD, Nokia, ...) (044, 055) Isabelle Amonou CE6: Improvement of update step (015, 030, 062) (Samsung, MSRA, Nokia, FhG-HHI) Woo-Jin Han CE7: Enhancement-layer intra prediction (Thomson, FhG-HHI, Sharp, Huawei, Samsung) (010, 053, 065) Jill Boyce CE8: Region of Interest (NCTU, ICU, ETRI, I2R) (020) Zhongkang Lu CE9: Improvement of quantization (046, 060, 066, 069) (FTRD, Panasonic, Siemens, RWTH, FhG-HHI, Microsoft, Sharp) St?phane Pateux CE10: Extended spatial scalability (Thomson, FTRD, Sharp, LG) (008, 041, 042) Edouard Francois CE11: Improvement of FGS (055) (Nokia, FhG-HHI, NCTU) Justin Ridge CE12: Weighted prediction from FGS layers (054) (Nokia, Visiowave, FhG-HHI) Yiliang Bao On the ISO/IEC side, the standards developed by the JVT are published as part of the ISO/IEC 14496 suite of standards, which is available for purchase at: http://tinyurl.com/2dgpx The primary such standard, 14496-10 (2023-09-28) "Information technology -- Coding of audio-visual objects -- Part 10: Advanced Video Coding", is available for purchase at: http://tinyurl.com/6dnck On the ITU-T side, the standards developed by the JVT are published as part of the "H-series" of standards. Anyone can get copies of any 3 ITU-T standards per year FOR FREE by using the following link: http://ecs.itu.ch/cgi-bin/ebookshop The links to the JVT's standards at ITU-T are as follows: ------------------------------------------------------------- Title: H.264 (2005-03): "Advanced video coding for generic audiovisual services" URL: http://tinyurl.com/62t46 ------------------------------------------------------------- Title: H.264.1 (2005-03): "Conformance specification for H.264 advanced video coding" URL: http://tinyurl.com/5qp7g ------------------------------------------------------------- Title: H.264.2 (2005-03): "Reference software for H.264 advanced video coding" URL: http://tinyurl.com/6flcp ------------------------------------------------------------- Best regards, Gary Sullivan From cho8629 dreamwiz.com Thu May 26 11:42:40 2005 From: cho8629 dreamwiz.com (Ik-Hwan Cho) Date: Thu May 26 03:50:40 2005 Subject: [Mp4-tech] MPEG-2 TS converter Message-ID: <200505260142.j4Q1gkYA051353@smtp0.dreamwiz.com> Dear experts.. I will have a plan to test simulate MPEG-2 TS streaming service. I get source code of mpeg-2 encoder from http://www.mpeg.org , but I can?t convert the encoded mpeg-2 file to mpeg-2 TS stream.. Is there anyone who lets me know how to convert mpeg-2 file to mpeg-2 TS stream.. ___________________________________________________________________________ ??? (Ik-Hwan Cho) 402-751 ??? ?? ??3? ????? ????? ????? ???(2N575) 2N575, Multimedia Lab. Electronic Engineering, Inha University Yong Hyun 3 Dong, NamGu, Incheon, 402-751, South Korea E-mail : cho8629@dreamwiz.com MSN : cho8629@korea.com Tel : +82-32-860-7415 Mobile : +82-018-267-4535 __________________________________________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050526/77ca1692/attachment-0001.html From tuomo.toukomies gmail.com Thu May 26 10:15:34 2005 From: tuomo.toukomies gmail.com (Tuomo Toukomies) Date: Thu May 26 03:50:46 2005 Subject: [Mp4-tech] Is there a way to log the actual framerate of a video clip over time? In-Reply-To: References: Message-ID: Hi. Does anyone know of any video analysis SW which could log the actual framerate of a video clip over time? The framerate of very low (mobile) bitrate .mp4 and .3gp videos I make is typically lower than the target I set: i.e. for 20kbps Mpeg-4 QCIF video at 15fps, I typically get 3-6 fps in the actual encoded file. This is normal as far as I understand, as the low bitrate does not allow for more. (I use various tools to make the clips, including Quicktime Pro, Real Mobile Producer and Sorenson) I need to compare the actual framerate in the encoded video to the playback speed achieved on a mobile device. I can log the framerate achieved in the mobile device, but I have found no way to "profile" the actual framerate over time in the original video. What I need to get is a log file stating how many frames of actual video there are for each second in the encoded video clip. I know I could create such a profile by hand using some video player that allows me to play the video frame by frame, but I would very much like to automate this process. Any ideas? Thank you, Tuomo Toukomies -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050526/66870b15/attachment.html From alexismt comcast.net Thu May 26 03:20:04 2005 From: alexismt comcast.net (Alexis Michael Tourapis) Date: Thu May 26 07:19:07 2005 Subject: [Mp4-tech][Video H.264] Error in one quarter interpolation for lumasamples In-Reply-To: <84ac0130505250745512691d9@mail.gmail.com> Message-ID: <200505260920.j4Q9KBnc016727@lists1.magma.ca> Dear Alejandro, You may notice that j1 is computed as: j1 = cc - 5 * dd + 20 * h1 + 20 * m1 - 5 * ee + ff, or (8-241) j1 = aa - 5 * bb + 20 * b1 + 20 * s1 - 5 * gg + hh (8-242) Note that aa-hh and h1,b1,s1, and n1 are derived by only applying the 6 tap and without considering the right shift operation. This is also mentioned in the text after these two equations, i.e.: ================= where intermediate values denoted as aa, bb, gg, s1 and hh shall be derived by applying the 6-tap filter horizontally in the same manner as the derivation of b1 and intermediate values denoted as cc, dd, ee, m1 and ff shall be derived by applying the 6-tap filter vertically in the same manner as the derivation of h1. ================== Best regards, Alexis -----Original Message----- From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Alejandro Moya Sent: Wednesday, May 25, 2024 7:46 AM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech][Video H.264] Error in one quarter interpolation for lumasamples Hi, everyone I have studying H.264 standard lately and I discovered yesterday what I guess it could be an error: when you apply the 6-tap filter to get the pixel labeled as "j" in Figure 8-4, the formulae is identical to the one you have to apply to get "b" or "h" with one huge difference: for obtaining the last ones you do b = Clip1( (b1 + 16)>>5 ) (8-187) h = Clip1( (h1 + 16)>>5 ) (8-188) and to obtain "j" you do: j = Clip1((j1 + 512)>>10) This means that in the first case the filter is modulo 1, but in the second is modulo 1/32. I think the proper formulae to get "j" should be: j = Clip1( (j1 + 16)>>5 ) (8-191) Let's say that full pel samples G, H, M and N (the surrounding ones to "j") have all the value 0xB2, so common sense tells us that the luma pixel at 1/4 position (the one called "j") should be 0xB2 too. But applying equation (8-191) we get the value ox 0x06 at that 1/4 interpolated position, so it makes no sense. I displayed the interpolated image and got some ugly "black dots" at that interpolated positions. I would like to ask all of you what do you think about that. Maybe I am missing some littled detail when interpolating "j" or maybe standard is wrong? I have the last revision which I just downloaded today from ITU-T webpage. Thanks you!! _______________________________________________ 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 B.McAllister ee.qub.ac.uk Thu May 26 11:57:30 2005 From: B.McAllister ee.qub.ac.uk (Brendan McAllister) Date: Thu May 26 07:22:31 2005 Subject: [Mp4-tech] PSNR software Message-ID: Hi Experts, Could someone please point me in the direction of where I could get PSNR calculation software, preferably freeware. Brendan McAllister -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050526/435cd26c/attachment.html From almomo1 gmail.com Thu May 26 13:06:55 2005 From: almomo1 gmail.com (Alejandro Moya) Date: Thu May 26 07:24:15 2005 Subject: [Mp4-tech][Video H.264] Error in one quarter interpolation for lumasamples In-Reply-To: <429594cc.0e6e27f1.6768.ffffdc1bSMTPIN_ADDED@mx.gmail.com> References: <84ac0130505250745512691d9@mail.gmail.com> <429594cc.0e6e27f1.6768.ffffdc1bSMTPIN_ADDED@mx.gmail.com> Message-ID: <84ac01305052603066371411b@mail.gmail.com> Thank you very much!! I didn't notice that little big detail :D On 5/26/05, Alexis Michael Tourapis wrote: > Dear Alejandro, > > You may notice that j1 is computed as: > > j1 = cc - 5 * dd + 20 * h1 + 20 * m1 - 5 * ee + ff, or (8-241) > j1 = aa - 5 * bb + 20 * b1 + 20 * s1 - 5 * gg + hh (8-242) > > Note that aa-hh and h1,b1,s1, and n1 are derived by only applying the 6 tap > and without considering the right shift operation. This is also mentioned in > the text after these two equations, i.e.: > > ================= > where intermediate values denoted as aa, bb, gg, s1 and hh shall be > derived by applying the 6-tap filter horizontally in the same manner as the > derivation of b1 and intermediate values denoted as cc, dd, ee, m1 and ff > shall be derived by applying the 6-tap filter vertically in the same manner > as the derivation of h1. > ================== > > Best regards, > > Alexis > > -----Original Message----- > From: mp4-tech-bounces@lists.mpegif.org > [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Alejandro Moya > Sent: Wednesday, May 25, 2024 7:46 AM > To: mp4-tech@lists.mpegif.org > Subject: [Mp4-tech][Video H.264] Error in one quarter interpolation for > lumasamples > > Hi, everyone > > I have studying H.264 standard lately and I discovered yesterday what > I guess it could be an error: when you apply the 6-tap filter to get > the pixel labeled as "j" in Figure 8-4, the formulae is identical to > the one you have to apply to get "b" or "h" with one huge difference: > for obtaining the last ones you do > b = Clip1( (b1 + 16)>>5 ) (8-187) > h = Clip1( (h1 + 16)>>5 ) (8-188) > and to obtain "j" you do: > j = Clip1((j1 + 512)>>10) > > This means that in the first case the filter is modulo 1, but in the > second is modulo 1/32. I think the proper formulae to get "j" should > be: > j = Clip1( (j1 + 16)>>5 ) (8-191) > > Let's say that full pel samples G, H, M and N (the surrounding ones to > "j") have all the value 0xB2, so common sense tells us that the luma > pixel at 1/4 position (the one called "j") should be 0xB2 too. But > applying equation (8-191) we get the value ox 0x06 at that 1/4 > interpolated position, so it makes no sense. I displayed the > interpolated image and got some ugly "black dots" at that interpolated > positions. > > I would like to ask all of you what do you think about that. Maybe I > am missing some littled detail when interpolating "j" or maybe > standard is wrong? I have the last revision which I just downloaded > today from ITU-T webpage. > > > Thanks you!! > > _______________________________________________ > 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 almomo1 gmail.com Thu May 26 13:27:59 2005 From: almomo1 gmail.com (Alejandro Moya) Date: Thu May 26 07:26:10 2005 Subject: [Mp4-tech] MPEG-2 TS converter In-Reply-To: <200505260142.j4Q1gkYA051353@smtp0.dreamwiz.com> References: <200505260142.j4Q1gkYA051353@smtp0.dreamwiz.com> Message-ID: <84ac013050526032723a69ced@mail.gmail.com> Dear Ik.Hwan Cho, converting MPEG-2 Program stream into one Transport Stream could notbe difficult if you just to mux one program into the TS without timinginformation: you should remove Pack and System Headers from ProgramStream and should divide the resulting Packets into 184 byte chunksadding an easy 4 byte header -> 0x47 0x0Y 0xYY 0x10 where "Y" are 13bits indicating your Program IDentification (you may choose it as youdesire). The last step would be to add two packets at the beginning ofthe file containing your new TS: one with Program Association Table(Tranport Stream which should have PID = 0) and Program Map Table(which will have the PID you defined in the Program Association Table)containing the PIDs of the transport stream packets carrying youroriginal video (i.e., the PID "Y" you choose above). Instead, you could try "Moonlight MPEG-2 Encoder + Streaming Pack"which hopefully will allow you to encode directly into a TS; theproblem is that you will have to pay for it. Best regards, Alejandro On 5/26/05, Ik-Hwan Cho wrote:> > > Dear experts..> > > > I will have a plan to test simulate MPEG-2 TS streaming service. I get> source code of mpeg-2 encoder from http://www.mpeg.org , but I can't convert> the encoded mpeg-2 file to mpeg-2 TS stream..> > Is there anyone who lets me know how to convert mpeg-2 file to mpeg-2 TS> stream..> > > > > > ___________________________________________________________________________> ??? (Ik-Hwan Cho) > 402-751 ??? ?? ??3? ????? ????? ????? ???(2N575) > 2N575, Multimedia Lab. Electronic Engineering, Inha University > Yong Hyun 3 Dong, NamGu, Incheon, 402-751, South Korea > E-mail : cho8629@dreamwiz.com > MSN : cho8629@korea.com > Tel : +82-32-860-7415 > Mobile : +82-018-267-4535 > __________________________________________________________________________> > > _______________________________________________> 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 sathishn teneoris.com Thu May 26 12:39:56 2005 From: sathishn teneoris.com (Sathish N) Date: Sat May 28 18:18:12 2005 Subject: [Mp4-tech] Conformance testing of MPEG4 Advanced Simple profile (Video) Message-ID: <1117107613.2175.13.camel@sathish> Hello, I'm looking for Conformance tests & bitstreams for MPEG-4 video for Advanced Simple profile. In this process, I purchased ISO Conformance document (ISO/IEC 14496-4) & the test-streams. 1. I found 13 bitstreams under the directory 'advanced_simple'. But the Conformance document does not have any description about these bitstreams. No Table given in the document has any mention on 'Advanced Simple profile'. 2. I came across a link : 'MPEG-4 Video Conformance Bitstreams available for download' in MPEG4 Industry forum (m4if.org). Under this link also, I found some bit-streams under the directory 'Advanced_Simple'. These streams are different from the ones which are distributed from ISO. Also, these appear to be newer than the ones from ISO (as evident from the timestamps of the files). 3. Eventhough ISO/IEC 14496-4 document does not seem to categorize any bitstreams under Advanced Simple profile, there are some bit-streams listed under Section 5.6 - 'Additional Conformance Testing' which have the features of Advanced Simple profile like Quarter-sample interpolation & GMC. In view of the above, I'm not quite clear about which of the sources I should be using (1, 2 or 3 mentioned above) for testing the features of my Advanced Simple profile decoder. Would some one please clarify ? Regards, Sathish From tuukkat ee.oulu.fi Thu May 26 16:05:52 2005 From: tuukkat ee.oulu.fi (Tuukka Toivonen) Date: Sat May 28 18:22:37 2005 Subject: [Mp4-tech] PSNR software In-Reply-To: References: Message-ID: On Thu, 26 May 2005, Brendan McAllister wrote: > Could someone please point me in the direction of where I could get PSNR > calculation software, preferably freeware. You don't tell from what you want to calculate PSNRs. Anyway, I suppose I can put my tools online, although they aren't very polished. http://www.ee.oulu.fi/%7Etuukkat/tmp/yuvtools-0.1.tar.gz There are many small utilities included, which mainly handle YUV files (or AVIs and Quicktime files containing images in the same format). yuvcmp is what you're asking for; it computes the PSNR between two YUV files, like this: > ./yuvcmp -1 /tmp/waterfall-352x288.yuv -2 /video/std-video/waterfall-352x288.yuv Comparing size 352x288 Frame 0 ssd=94793193 ssd_y=15849597 ssd_u=43241511 ssd_v=35702085 Frame 1 ssd=95025238 ssd_y=16072043 ssd_u=43323994 ssd_v=35629201 Frame 2 ssd=95253675 ssd_y=16130628 ssd_u=43481072 ssd_v=35641975 Frame 3 ssd=95339443 ssd_y=16384598 ssd_u=43466749 ssd_v=35488096 ... Frame 98 ssd=102484627 ssd_y=20229242 ssd_u=45032018 ssd_v=37223367 Frame 99 ssd=103009597 ssd_y=20776749 ssd_u=45066791 ssd_v=37166057 Total: frames=100 psnr = 19.9639 psnr_y = 25.4078 psnr_u = 15.6907 psnr_v = 16.5733 avg_psnr = 19.9646 avg_psnr_y = 25.4147 avg_psnr_u = 15.6909 avg_psnr_v = 16.5736 As you can see, it computes both the average PSNR and the "correct" PSNRs for all color components separately as well as together. You probably need to edit Makefile to get it to compile. On my Linux box it compiles easily, but some of the programs require libquicktime package installed. From tomar_atul2000 yahoo.com Thu May 26 08:04:11 2005 From: tomar_atul2000 yahoo.com (atul tomar) Date: Sat May 28 18:25:12 2005 Subject: [Mp4-tech] Help me H.264.... Message-ID: <20050526140411.86144.qmail@web52306.mail.yahoo.com> Dear Experts, I am new to h.264 standard, till now I tried to read NAL unit sementics, RBSP sementics and slice header sementics from DRAFT ITU-T Rec. H.264 (2002 E). I have lot of doubts, Can anybody please let me know whether I am going in a right direction. I would appreciate if anybody tell me how should I proceed to implement h.264 baseline profile.How should I move forward in my project. Am I following updated current standard draft? Where is current h.264 standard doc(free) to follow, can someone locate it. Thanks in advance. Reg, Atul __________________________________________________ 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/20050526/11267560/attachment.html From princeofpersia79 gmail.com Thu May 26 14:39:12 2005 From: princeofpersia79 gmail.com (viraj) Date: Sat May 28 18:28:05 2005 Subject: [Mp4-tech] Data Partitioning in H.264 Message-ID: <77f040c105052610391f2a00cc@mail.gmail.com> Hi, I am trying to get multiple bitstreams using the data partitioning of H.264 (JM96). Even if I set that option in the encoder.cfg file (DP = 1), I am unable to get three data partitions (or three bit streams). All I get is a single bitstream (test.264). I have following questions: 1. Is it possible to obtain multiple H.264 bitstreams in mulitple files using data partitioning? 2. Does the data partitioning option provided in the encoder.cfg config file refer to multiple bitstreams or do I misunderstand the option? 3. Is there any other JM version of H.264 that implements data partitioning to obtain mulitple bitstreams? Thank you From princeofpersia79 gmail.com Thu May 26 22:00:54 2005 From: princeofpersia79 gmail.com (viraj) Date: Sat May 28 18:30:13 2005 Subject: [Mp4-tech] Data Partitioning in H.264 Message-ID: <77f040c1050526180014962021@mail.gmail.com> Hi, I am trying to get multiple bitstreams using the data partitioning of H.264 (JM96). Even if I set that option in the encoder.cfg file (DP = 1), I am unable to get three data partitions (or three bit streams). All I get is a single bitstream (test.264). I have following questions: 1. Is it possible to obtain multiple H.264 bitstreams in mulitple files using data partitioning? 2. Does the data partitioning option provided in the encoder.cfg config file refer to multiple bitstreams or do I misunderstand the option? 3. Is there any other JM version of H.264 that implements data partitioning to obtain mulitple bitstreams? Thank you From bharatsoni gmail.com Fri May 27 11:47:33 2005 From: bharatsoni gmail.com (Bharat Soni) Date: Sat May 28 18:33:05 2005 Subject: [Mp4-tech] Is there a way to log the actual framerate of a video clip over time? In-Reply-To: References: Message-ID: Hi, There are quite a few companies that are into development of stream analyzer software; Venera Technologies (www.venera.co.in), Vqual etc. You may get the required information from these companies. Venera also provides a freeware that provides high level stream statistics. May that can be of help to you. Regards, Bharat On 5/26/05, Tuomo Toukomies wrote: > Hi. > > > Does anyone know of any video analysis SW which could log the actual > framerate of a video clip over time? > > The framerate of very low (mobile) bitrate .mp4 and .3gp videos I make is > typically lower than the target I set: i.e. for 20kbps Mpeg-4 QCIF video at > 15fps, I typically get 3-6 fps in the actual encoded file. This is normal as > far as I understand, as the low bitrate does not allow for more. (I use > various tools to make the clips, including Quicktime Pro, Real Mobile > Producer and Sorenson) > > I need to compare the actual framerate in the encoded video to the playback > speed achieved on a mobile device. I can log the framerate achieved in the > mobile device, but I have found no way to "profile" the actual framerate > over time in the original video. > > What I need to get is a log file stating how many frames of actual video > there are for each second in the encoded video clip. I know I could create > such a profile by hand using some video player that allows me to play the > video frame by frame, but I would very much like to automate this process. > > Any ideas? > > Thank you, > > > Tuomo Toukomies > _______________________________________________ > 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 janisjp hanmail.net Fri May 27 15:54:33 2005 From: janisjp hanmail.net (janis) Date: Sat May 28 18:35:47 2005 Subject: [Mp4-tech] H.264/AVC Reference Software Encoder Documentation Message-ID: <20050527145433.HM.00000000000MPwZ@wwl27.hanmail.net> An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050527/88a6db92/attachment.html From reply2manoj gmail.com Fri May 27 16:03:21 2005 From: reply2manoj gmail.com (manoj t) Date: Sat May 28 18:39:02 2005 Subject: [Mp4-tech] Latest AAC SBR conformance tool Message-ID: <1f8628de05052702334d73413f@mail.gmail.com> Hi, where can I get the latest AAC SBR conformance tool .exe or source code? Thanks, Manoj From Nicolas.GAULLIER CORP.TPS.fr Fri May 27 16:41:34 2005 From: Nicolas.GAULLIER CORP.TPS.fr (Nicolas.GAULLIER@CORP.TPS.fr) Date: Sat May 28 18:42:01 2005 Subject: [Mp4-tech] bug in reference decoder : pic_struct is 3 bits / 4 bits in the spec Message-ID: Hi, I use the reference decoder to analyse h264 streams and parse/display SEI messages. I had some unexpected/incorrect values for picture structure and ct_type (pic_struc=1=top field and ct_type=0=progressive), so I checked the source code -/- the specs : * reference decoder - "sei.c" picture_structure = u_v(3, "SEI: picture_structure" , buf); * specs 14496-10 - D.1.2 "picture timing SEI message syntax" pic_struct u(4) I change the line in sei.c to picture_structure = u_v(4, "SEI: picture_structure" , buf); And now the values I get seem meaningful (I have a pic_struct=3 and no ct_type). (Note: this is in last JM96, but it was already in older JM versions) Best regards, Nicolas Gaullier > ------------------------------------------------------------------------------------- > Nicolas Gaullier > T?l?vision Par Satellite > D?partement Broadcast et Produits > Service Ing?nierie > 145 Quai de Stalingrad > 92137 ISSY LES MOULINEAUX CEDEX > FRANCE > * +33 1 41 33 28 24 > ----------------------------------------------------------------------------------------------- Ce message electronique et tous les fichiers joints ainsi que les informations contenues dans ce message ( ci apres "le message" ), sont confidentiels et destines exclusivement a l'usage de la personne a laquelle ils sont adresses. Si vous avez re?u ce message par erreur, merci de le renvoyer a son emetteur et de le detruire. Toutes diffusion, publication, totale ou partielle ou divulgation sous quelque forme que ce soit non expressement autorisees de ce message, sont interdites. Les idees et opinions presentees dans ce message sont celles de son auteur, et ne representent pas necessairement celles de TPS (et/ou des entites membres de la societe TPS et de ses filiales). This e-mail, any attachments and the information contained there in ("this message") are confidential and intended solely for the use of the addressee(s). If you have received this message in error please send it back to the sender and delete it. Unauthorized publication, use, dissemination or disclosure of this message, either in whole or in part is strictly prohibited. Any views or opinions presented are solely those of its author and do not necessarily represent those of TPS (and/or its group companies). ----------------------------------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050527/9e8865a1/attachment-0001.html From gthm159 yahoo.co.in Sat May 28 11:29:46 2005 From: gthm159 yahoo.co.in (Googie) Date: Sat May 28 18:45:44 2005 Subject: [Mp4-tech] [video] Mpeg4 resync marker Message-ID: <20050528092946.61006.qmail@web53708.mail.yahoo.com> Hi, My name is Gautam and I am developing a software to encode and decode mpeg4 video streams. But right now I am stuck because I don't know the exact format of the mpeg4 resync marker. Please help me out. regards, Gautam my e-mail: gthm159@yahoo.co.in Yahoo! India Matrimony: Find your partner online. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050528/4d0a4a12/attachment.html From zhihangwang jdl.ac.cn Fri May 27 01:19:26 2005 From: zhihangwang jdl.ac.cn (=?gb2312?B?zfXWvrq9?=) Date: Sun May 29 09:53:30 2005 Subject: [Mp4-tech] Think how to debug me Message-ID: <002101c5620e$b0e3caf0$0e00a8c0@myntserver> Dear All Experts: Who would like to give some advice on debuging me? How to confirm the motion vection's correctness easyly? When I debug motion estimation, I have to step into almost every search point so that I can know whether the point is right. Best Regards -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050527/fc6d8876/attachment.html From dmitriy graphics.cs.msu.ru Sun May 29 15:00:57 2005 From: dmitriy graphics.cs.msu.ru (Dmitriy Vatolin) Date: Sun May 29 09:56:46 2005 Subject: [Mp4-tech] www.compression-links.info Message-ID: <1772470509.20050529140057@graphics.cs.msu.ru> Hello all! New compression sites catalog was started with support of MSU Graphics&Media Lab: http://www.compression-links.info/ This site was created as a live collection of freely submitted links to the data compression resources and to the resources on video, image, and audio processing as well. The original set of links was granted by Mark Nelson. We write site engine on Perl (+ Apache + MySQL + Linux) and now we check, update and add links with good speed. Now there: * 2400 links * 6000 visitors per month Mark Nelson did not add enough links on video and hope you can add more useful links to categories: http://www.compression-links.info/MPEG http://www.compression-links.info/Video http://www.compression-links.info/MPEG-4_AVC_H264 http://www.compression-links.info/MP3 If you find we omit some useful site - please let me know, I'l working on video & audio areas and update catalog soon. Or you can register and add link by yourself. Enjoy! ;) -- Best regards, Dmitriy mailto:dmitriy@graphics.cs.msu.ru From kexu ee.cuhk.edu.hk Sun May 29 21:24:28 2005 From: kexu ee.cuhk.edu.hk (Xu Ke) Date: Sun May 29 09:58:11 2005 Subject: [Mp4-tech] Help me H.264.... Message-ID: <200505291224.j4TCOKtL006100@cuees8.ee.cuhk.edu.hk> Hi Atul, I would suggest you the following sequence: 1)IEEE transactions on circuits and systems for video technology,July 2003,this is a special edition for H.264 2)The book "H.264 and MPEG-4 Video Compression" 3)After you have a brief feeling on H.264,you can start to read the standard and try to use JM software.I strongly suggest you to read the trace file dumped out by JM then you can find how each syntax is parsed in the bitstream. Best regards, XU Ke ======= 2024-05-26 22:04:11 ================= >Dear Experts, > >I am new to h.264 standard, till now I tried to read NAL unit sementics, RBSP sementics and slice header sementics from DRAFT ITU-T Rec. H.264 (2002 E). I have lot of doubts, Can anybody please let me know whether I am going in a right direction. I would appreciate if anybody tell me how should I proceed to implement h.264 baseline profile.How should I move forward in my project. >Am I following updated current standard draft? Where is current h.264 standard doc(free) to follow, can someone locate it. > >Thanks in advance. >Reg, >Atul > > >__________________________________________________ >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 = = = = = = = = = = = = = = = = = = = = From AngelZhang innosis.com.cn Mon May 30 10:32:18 2005 From: AngelZhang innosis.com.cn (AngelZhang@innosis.com.cn) Date: Mon May 30 04:19:02 2005 Subject: [Mp4-tech] H.264:about data of 2 frame QCIF yuv file Message-ID: hi,experts who can tell me how many data in a 2 frame yuv file in QCIF format? in my opinion,with the 4:2:0 basis of the propotion between Y U V, there should be 144+71 lines with 176 datas in each lines in one frame. then, there are 2*(144+71)*176 datas in the 2 frame yuv file. is there any wrong? please give me some hints. thanks in advance angel From ksuehring web.de Mon May 30 10:38:38 2005 From: ksuehring web.de (Karsten Suehring) Date: Mon May 30 04:22:41 2005 Subject: [Mp4-tech] bug in reference decoder : pic_struct is 3 bits / 4 bits in the spec In-Reply-To: References: Message-ID: <429AC2FE.5000908@web.de> Dear Nicolas, thank you for reporting the bug. I will fix that in the next release. Best regards, Karsten Nicolas.GAULLIER@corp.tps.fr wrote: > > > Hi, > > I use the reference decoder to analyse h264 streams and parse/display > SEI messages. I had some unexpected/incorrect values for picture > structure and ct_type (pic_struc=1=top field and ct_type=0=progressive), > so I checked the source code -/- the specs : > > * reference decoder - "sei.c" > picture_structure = u_v(3, "SEI: picture_structure" , buf); > * specs 14496-10 - D.1.2 "picture timing SEI message syntax" > pic_struct u(4) > > I change the line in sei.c to > picture_structure = u_v(4, "SEI: picture_structure" , buf); > > And now the values I get seem meaningful (I have a pic_struct=3 and no > ct_type). > (Note: this is in last JM96, but it was already in older JM versions) > > Best regards, > Nicolas Gaullier > ------------------------------------------------------------------------------------- > > Ni colas Gaullier > T?l?vision Par Satellite > D?partement Broadcast et Produits > Service Ing?nierie > 145 Quai de Stalingrad > 92137 ISSY LES MOULINEAUX CEDEX > FRANCE > ( +33 1 41 33 28 24 > > ----------------------------------------------------------------------------------------------- > Ce message electronique et tous les fichiers joints ainsi que les > informations contenues > dans ce message ( ci apres "le message" ), sont confidentiels et > destines exclusivement a > l'usage de la personne a laquelle ils sont adresses. Si vous avez re?u > ce message par erreur, > merci de le renvoyer a son emetteur et de le detruire. Toutes diffusion, > publication, totale > ou partielle ou divulgation sous quelque forme que ce soit non > expressement autorisees de ce > message, sont interdites. Les idees et opinions presentees dans ce > message sont celles de son > auteur, et ne representent pas necessairement celles de TPS (et/ou des > entites membres de la > societe TPS et de ses filiales). > > This e-mail, any attachments and the information contained there in > ("this message") are > confidential and intended solely for the use of the addressee(s). If you > have received this > message in error please send it back to the sender and delete it. > Unauthorized publication, > use, dissemination or disclosure of this message, either in whole or in > part is strictly > prohibited. Any views or opinions presented are solely those of its > author and do not > necessarily represent those of TPS (and/or its group companies). > ----------------------------------------------------------------------------------------------- > > > ------------------------------------------------------------------------ > > _______________________________________________ > 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 abdessamad.falhi epfl.ch Mon May 30 11:12:33 2005 From: abdessamad.falhi epfl.ch (abdessamad.falhi@epfl.ch) Date: Mon May 30 17:49:13 2005 Subject: [Mp4-tech] AAC from mp4 Message-ID: <1117440753.429acaf13a907@imapwww.epfl.ch> Hi, Does anyone have a tool to extract AAC features from an mp4 file? with regards? From tuukkat ee.oulu.fi Mon May 30 12:19:10 2005 From: tuukkat ee.oulu.fi (Tuukka Toivonen) Date: Mon May 30 17:49:19 2005 Subject: [Mp4-tech] Help me H.264.... In-Reply-To: <20050526140411.86144.qmail@web52306.mail.yahoo.com> References: <20050526140411.86144.qmail@web52306.mail.yahoo.com> Message-ID: On Thu, 26 May 2005, atul tomar wrote: > Am I following updated current standard draft? Where is current h.264 >standard doc(free) to follow, can someone locate it. You can download a few standards here for free: http://ecs.itu.ch/cgi-bin/ebookshop H.264 standard is there, though seems to be missing at least one picture... From jiazhaox mobilesoft.com.cn Mon May 30 17:22:14 2005 From: jiazhaox mobilesoft.com.cn (JazryXu) Date: Mon May 30 17:49:26 2005 Subject: [Mp4-tech] H.264:about data of 2 frame QCIF yuv file References: Message-ID: <009001c564f0$af889660$7ca1a8c0@JazryXu> the size is 176*144*1.5*2, what is the 71? ----- Original Message ----- From: To: Sent: Monday, May 30, 2024 9:32 AM Subject: [Mp4-tech] H.264:about data of 2 frame QCIF yuv file > hi,experts > who can tell me how many data in a 2 frame yuv file in QCIF format? in my > opinion,with the 4:2:0 basis of the propotion between Y U V, there should > be > 144+71 lines with 176 datas in each lines in one frame. then, there are > 2*(144+71)*176 datas in the 2 frame yuv file. > is there any wrong? please give me some hints. > > thanks in advance > angel > > > _______________________________________________ > 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 tuukkat ee.oulu.fi Mon May 30 12:40:12 2005 From: tuukkat ee.oulu.fi (Tuukka Toivonen) Date: Mon May 30 17:49:32 2005 Subject: [Mp4-tech] H.264:about data of 2 frame QCIF yuv file In-Reply-To: References: Message-ID: On Mon, 30 May 2024 AngelZhang@innosis.com.cn wrote: > who can tell me how many data in a 2 frame yuv file in QCIF format? in my > opinion,with the 4:2:0 basis of the propotion between Y U V, there should > be > 144+71 lines with 176 datas in each lines in one frame. then, there are > 2*(144+71)*176 datas in the 2 frame yuv file. > is there any wrong? please give me some hints. Each frame in a YUV file takes up W*H*1.5 bytes of memory, or Y (luma): W * H bytes U (chroma): W/2 * H/2 bytes V (chroma): W/2 * H/2 bytes For 2 frame file at QCIF resolution, 176*144*1.5 * 2 = 76032 bytes. From delyank gmail.com Mon May 30 15:18:10 2005 From: delyank gmail.com (Delyan Kalchev) Date: Mon May 30 17:49:38 2005 Subject: [Mp4-tech] LATM specifications Message-ID: I found this in SP-040656 from 3GPP /* /* The payload format RFC 3016 for MPEG-4 AAC specify that the audio streams shall be formatted by the LATM (Low-overhead MPEG-4 Audio Transport Multiplex) tool [21]. It should be noted that the references for the LATM format in the RFC 3016 [13] point to an older version of the LATM format than included in [21]. In [21] a corrigendum to the LATM tool is included. This corrigendum includes changes to the LATM format making implementations using the corrigendum incompatible with implementations not using it. To avoid future interoperability problems, implementations of PSS client and servers supporting AAC shall follow the changes to the LATM format included in [21]. */ AND [21] ISO/IEC 14496-3:2001: "Information technology ? Coding of audio-visual objects ? Part 3: Audio". RFC3016: [3] ISO/IEC 14496-3:1999, "Information technology - Coding of audiovisual objects - Part3: Audio". */ As far as I can see there are lots of standards, AMD.-s and COR.-s to 14496-3 We also have this: ISO/IEC 14496-3:2001/Cor 1:2002 to [21] and ISO/IEC 14496-3:2001/Cor 2:2004, and who knows what more. The questions are: 1. Are they really incompatible and which ones are or are not? 2. What makes them incompatible? 3. How can I realise something that would be universal? Actually I need to know what exactly standarts should I read. If I act according to ISO/IEC 14496-3:2001/Cor 2:2004 what would be compatible and what not? 4. Is LATM covered entirely in subclause 1.7 (MPEG-4 Audio transport stream)? As far as I can see it is entirely replaced in ISO/IEC 14496-3:2001/Cor 2:2004. Please be nice to direct me to other mailing lists if my questions are not to be asked here! Thanks Delyan Kalchev From hank_wallace hotmail.com Mon May 30 16:42:32 2005 From: hank_wallace hotmail.com (Hank Wallace) Date: Mon May 30 17:49:44 2005 Subject: [Mp4-tech] Is there a way to log the actual framerate of a video clip over t Message-ID: Hi Tuomo, For any kind of mpeg-4 analysis, you should also check out SceneScope (www.sceneware.nl). It's my favorite. Hank G. Wallace >On May 27, 2005, at 07:17, Bharat Soni wrote: >Hi, >There are quite a few companies that are into development of stream >analyzer software; Venera Technologies (www.venera.co.in), Vqual etc. >You may get the required information from these companies. Venera also >provides a freeware that provides high level stream statistics. May >that can be of help to you. >Regards, >Bharat >On 5/26/05, Tuomo Toukomies wrote: >Hi. >Does anyone know of any video analysis SW which could log the actual >framerate of a video clip over time? >The framerate of very low (mobile) bitrate .mp4 and .3gp videos I make is >typically lower than the target I set: i.e. for 20kbps Mpeg-4 QCIF video at >15fps, I typically get 3-6 fps in the actual encoded file. This is normal >as >far as I understand, as the low bitrate does not allow for more. (I use >various tools to make the clips, including Quicktime Pro, Real Mobile >Producer and Sorenson) >I need to compare the actual framerate in the encoded video to the playback >speed achieved on a mobile device. I can log the framerate achieved in the >mobile device, but I have found no way to "profile" the actual framerate >over time in the original video. >What I need to get is a log file stating how many frames of actual video >there are for each second in the encoded video clip. I know I could create >such a profile by hand using some video player that allows me to play the >video frame by frame, but I would very much like to automate this process. >Any ideas? >Thank you, >Tuomo Toukomies _______________________________________________ >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 _________________________________________________________________ From ranzhao arcsoft.com.cn Tue May 31 09:56:07 2005 From: ranzhao arcsoft.com.cn (Zhao Ran) Date: Tue May 31 10:58:59 2005 Subject: [Mp4-tech] Re: H.264:about data of 2 frame QCIF yuv file References: <200505301613.j4UG5kQZ024485@lists1.magma.ca> Message-ID: <000d01c5657b$87405450$6b7410ac@apac.arcsoft.corp> For every frame in QCIF format, the data size is 176*144(Y)+176*144/4(U)+176*144/4(V)=38016Byte > Date: Mon, 30 May 2024 09:32:18 +0800 > From: AngelZhang@innosis.com.cn > Subject: [Mp4-tech] H.264:about data of 2 frame QCIF yuv file > To: mp4-tech@lists.mpegif.org > Message-ID: > > > Content-Type: text/plain; charset=us-ascii > > hi,experts > who can tell me how many data in a 2 frame yuv file in QCIF format? in my > opinion,with the 4:2:0 basis of the propotion between Y U V, there should > be > 144+71 lines with 176 datas in each lines in one frame. then, there are > 2*(144+71)*176 datas in the 2 frame yuv file. > is there any wrong? please give me some hints. > > thanks in advance > angel From AngelZhang innosis.com.cn Tue May 31 10:33:47 2005 From: AngelZhang innosis.com.cn (AngelZhang@innosis.com.cn) Date: Tue May 31 11:06:36 2005 Subject: [Mp4-tech] Re: H.264:about data of 2 frame QCIF yuv file Message-ID: Hi,Thanks for the help of Zhao Ran,zhang.FanP,Dan searles,Ma zhan,Tuukka,Rajuesh,Jazry,Zhang.YingS. and first,let me apologize for the mistake of "71" which should have been spelled as "72". then with the format,we can get 432*176 bytes (here,432 means line number.)in a 2 frame YUV file. but what makes me annoied is that the output YUV file of H.264 decoder has only 244*176 bytes data. and in my opinion, even though some MB can be skipped in the process of encoder,but the decoder should add all data into the output *.yuv file. so please give me further hints. thanks very much! angel From dsn2603 rediffmail.com Tue May 31 06:16:41 2005 From: dsn2603 rediffmail.com (sakthi narayanan) Date: Tue May 31 11:06:42 2005 Subject: [Mp4-tech] AAC from mp4 Message-ID: <20050531051814.6584.qmail@webmail52.rediffmail.com> ? Hi, You can download the free mp4UI tool from this site http://www.free-codecs.com/mp4UI_download.htm It will extract the audio (aac raw bitstreams and adts aac bitstreams from mp4 file.U can convert the mp4 file to raw aac file. I think,it will be so useful for you. With regards sakthi On Mon, 30 May 2024 abdessamad.falhi@epfl.ch wrote : > > > Hi, > > Does anyone have a tool to extract AAC features from an mp4 file? > > with regards? >_______________________________________________ >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/20050531/de9283bd/attachment.html From garysull windows.microsoft.com Mon May 30 23:29:17 2005 From: garysull windows.microsoft.com (Gary Sullivan) Date: Tue May 31 11:06:50 2005 Subject: [Mp4-tech] H264 DPB size for practical decoders Message-ID: <91D7F2CEE3425A4A9D11311D09FCE2460EC40291@WIN-MSG-10.wingroup.windeploy.ntdev.microsoft.com> Ernest et al, You need to study Annex A (subclauses A.3.1 and A.3.2) and max_dec_frame_buffering in Annex E. Whether it is for purposes of referencing or display, all frames must fit into the DPB capacity for the profile and level. The constraint you're looking for is set by max_dec_frame_buffering, not num_reorder_frames or num_ref_frames. Note that max_dec_frame_buffering cannot be any larger than MaxDpbSize, which cannot be larger than 16 frames. First you should be aware that the MaxDpbSize is not as big as 16 frames when the individual frames are large. When operating at the maximum picture size for the profile and level, typically only 4-5 frames of capacity are needed. The upper limit of 16 is just for when the decoder is operating with pictures that are relatively small. Some minor amount of additional buffering (e.g., 1-3 frames) may also be recommended to cover some practical differences between an ideal hypothetical decoder and a real decoder. Best Regards, Gary Sullivan ________________________________ From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Ernest Lim Sent: Tuesday, May 10, 2024 11:30 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] H264 DPB size for practical decoders Hello all, I have a question regarding the size of the DPB and how this may be affected by the POC of decoded frames. I mailed this to JVT-EXPERTS but got no reply. Presumably on a practical system, possibly real-time, where for every period in which a frame is decoded, a frame is also displayed. Then if POC is not ordered in the same manner as frame number, there has to be a delay after the decoding commences before frames can be displayed. As I understand, this delay is determined by num_reorder_frames, which has a default value of 16 and is specified in the HRD parameter set. Do the decoded frames that are awaiting display, that are not being used for reference, reside in the DPB? This is what seems to happen in JM, although JM is able to "display" many frames in a single period so that when the DPB is filled with these frames awaiting display and reference frames, space is created by displaying and removing the next non-reference frame according to POC. What is supposed to happen if only 1 frame can be removed in a decoding period? My feeling is that the DPB should in fact be twice the size in this situation. To illustrate this, consider the following example. In this case, the DPB size is 16, num_reorder_frames is 16 and num_ref_frames is 16. The POCs of the following frames, listed in decoding order is: 0,2,...,15,16,17,1,18,19,...,31,32 The frame with POC 1 has 16 frames that precede it in decode order (2,...,17) but come after it in display order - this is consistent with num_reorder_frames of 16. Suppose that the 16 frames 0,1,2,...,15 are all used as reference frames for the entirety of the decode process once all are decoded and available. We would get the following sequence: 0(.),2(.),...,15(.),16(.),17(0),1(1),18(2),19(3),...,31(15),32(16) where the first number is the POC of the decoded frame and the one in brackets is the POC of the displayed frame. Since we are displaying 1 frame every decode period, the first frame that can be removed from the DPB is after the one with POC of 16 has been displayed. None of 0,1,...,15 can be removed since they are all being used as reference. POC 16 can be removed since it has been displayed and is not being used for reference. However, by this stage we have frames with POCs 0,...,31 in the DPB and are presumably about to insert the most recently decoded frame POC 32 into it! Is this situation correct, or have I misinterpreted something? It suggests that practical decoders that wish to display one frame every period in the correct display order will need a DPB of at least 2 times that stated in the spec. Sorry about the lengthy message. Regards, Ernie. ________________________________ Find local movie times and trailers on Yahoo! Movies. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050530/4bdec3c8/attachment-0001.html From shinjan.tiwary ittiam.com Tue May 31 12:19:50 2005 From: shinjan.tiwary ittiam.com (Shinjan Tiwary) Date: Tue May 31 11:06:57 2005 Subject: [Mp4-tech] regarding YUV Message-ID: <45A1F95BB9D7D84FAB0A1EB4D67EEF960196137D@is01ex01.ittiam.com> If you know the number of frames available in the sequence, you can calculate the total size of the file assuming 444, 422 and 420 formats and see which one matches Shinjan ------------------------------------------------------------------------------ Shinjan Kumar Tiwary Engineer, Audio-Speech Ittiam Systems (P) Ltd. Consulate 1, 1 Richmond Rd. Bangalore - 560025, INDIA +918056601129 (O) +919880016808 (M) _____________________________________________ -----Original Message----- From: amarnath nathan [mailto:lucyamarnath@yahoo.com] Sent: Monday, May 23, 2024 11:16 PM To: mp4-tech@lists.mpegif.org Subject: [Mp4-tech] regarding YUV hello experts would u please tell me how to find the format of a yuv file by looking the file. i.e how to know whether it is 4:4:4 or 4:2:2 or 4:2:0.since i have a file in a format (qcif_15fps.yuv). Also plz suggest some ideas to convert .YUV to .SEQ. regards amar _____ Discover Yahoo! Have fun online with music videos, cool games, IM & more. Check it out! -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050531/8fed4014/attachment.html From agulati neomagic.com Tue May 31 12:53:17 2005 From: agulati neomagic.com (Atul Gulati) Date: Tue May 31 11:07:05 2005 Subject: [Mp4-tech] AAC from mp4 In-Reply-To: <1117440753.429acaf13a907@imapwww.epfl.ch> References: <1117440753.429acaf13a907@imapwww.epfl.ch> Message-ID: <429C02D5.8000809@neomagic.com> Hi, You can use mp4ui tool, which allow to extract the AAC stream from the mp4 file.(you can save that as a .aac file) Regards, Atul. abdessamad.falhi@epfl.ch wrote: > Hi, > > Does anyone have a tool to extract AAC features from an mp4 file? > > with regards? >_______________________________________________ >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 abdessamad.falhi epfl.ch Tue May 31 10:45:23 2005 From: abdessamad.falhi epfl.ch (abdessamad.falhi@epfl.ch) Date: Tue May 31 11:07:13 2005 Subject: [Mp4-tech] AAC decoder channels Message-ID: <1117525523.429c1613d45f3@imapwww.epfl.ch> Hi, What is the idea behind implementing 48 channels in an AAC decoder? Can we restrict the number of channels to 2 channels if we want to create only a stereo decoder? with regards Abdessamad From zaidee rndtm.net.my Tue May 31 17:24:20 2005 From: zaidee rndtm.net.my (Ahmad Zaidee bin Abu) Date: Tue May 31 11:07:20 2005 Subject: [Mp4-tech] A doubt in calculation of PSNR value for Video Message-ID: <558E65614165F242AC17B5D316BFAB0F105628@mstmrndmx01.tmrnd.net.my> I believe that there is no way that MSE going to be 0. Unless if the compressed frame is identical to original image. Then why do we care calculate PSNR if it is totally the same thing? Am i correct ? ________________________________ From: mp4-tech-bounces@lists.mpegif.org on behalf of M Mahdi Ghandi Sent: Thu 5/26/2005 1:12 AM To: 'Tuukka Toivonen' Cc: mp4-tech@lists.mpegif.org Subject: RE: [Mp4-tech] A doubt in calculation of PSNR value for Video The main problem is when for example this happened for only one frame in a sequence. I will not include that infinite in average PSNR calculation or alternatively replace it with an assumed upper limit (e.g. 50 dB). -----Original Message----- From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Tuukka Toivonen Sent: 25 May 2024 08:31 To: T. Bheemarjuna Reddy Cc: mp4-tech@lists.mpegif.org Subject: Re: [Mp4-tech] A doubt in calculation of PSNR value for Video On Tue, 24 May 2005, T. Bheemarjuna Reddy wrote: > I have one basic doubt in calculating PSNR value. > > PSNR (db) = 10 log {(2^n - 1)^2 / MSE} > > what happens to value of PSNR if MSE = 0 ? The PSNR is infinite then. You could check specifically for this case and return the largest possible number. If using IEEE floating points, they can represent infinite exactly. However, normally this is very rare case so people often ignore it. _______________________________________________ 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 The information contained in this Internet message is confidential and intended only for the use of the individual or entity identified. If the reader of this message is not the intended recipient, any dissemination, distribution or copying of the information contained in this Internet message is strictly prohibited. If you received this message in error, please notify the sender immediately. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050531/10ac81ed/attachment.html From dsn2603 rediffmail.com Tue May 31 11:41:25 2005 From: dsn2603 rediffmail.com (sakthi narayanan) Date: Tue May 31 11:07:29 2005 Subject: [Mp4-tech] AAC from mp4 Message-ID: <20050531104301.9084.qmail@webmail46.rediffmail.com> Hi, You can download the free mp4UI tool from this site http://www.free-codecs.com/mp4UI_download.htm It will extract the aac raw bitstreams from mp4 file.Using the same tool u can also extract adts aac bitstreams from mp4 file I think,it will be so useful for you. With regards sakthi On Mon, 30 May 2024 abdessamad.falhi@epfl.ch wrote : > > > Hi, > > Does anyone have a tool to extract AAC features from an mp4 file? > > with regards? >_______________________________________________ >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/20050531/4ef683cd/attachment.html From tomar_atul2000 yahoo.com Tue May 31 09:31:21 2005 From: tomar_atul2000 yahoo.com (atul tomar) Date: Tue May 31 16:00:12 2005 Subject: [Mp4-tech] about h.264 doubts.. Message-ID: <20050531153121.59955.qmail@web52304.mail.yahoo.com> Hello experts, I am new to h.264 standard but working very hard to understand the ITU standard. I have some doubts even after putting lot of thought process I could not understand the process. I would really appreciate if I could get your inputs on that. My doubts are: a) I am not able to understand how to determine frame_num and PreRefFrameNum in coded video sequence. Suppose I have frame sequence like I(IDR),P,P,P,P,I(IDR),P,P,P,P then how to get these two values for this sequence. I (rame_num,PreRefFrameNum),P(frame_num,PreRefFrameNum), P(frame_num,PreRefFrameNum),P(frame_num,PreRefFrameNum),P(frame_num,P reRefFrameNum), p(fame_num,PreRefFrameNum),I(frame_num,PreRefFrameNum),P(frame_num,PreRefFrameNum),P(frame_num,PreRefFrameNum),P(frame_num,PreRefFrameNum),p (frame_num,PreRefFrameNum). I reviewd 7.4.3 slice header sementics but could not get. b) As we know that we dont use interlacing in baseline proffile so Is it necessary to determine PictureOrderCount. I knew through standard that it is also used for decoder conformance checking but how Can we get TopField and BottomField field related variables if we are not dividing frame in fields then how can we work with above two variable while determing POC. As per my understanding, Is it correct if we consider only TopField variable thinking like a frame. Please give your inputs considering me as a newbie in this field. Many thanks in advance. Regards, Atul __________________________________________________ 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/20050531/14ed4b94/attachment.html From Gilberto ensilica.com Tue May 31 18:21:36 2005 From: Gilberto ensilica.com (Gilberto Rodriguez) Date: Tue May 31 16:07:43 2005 Subject: [Mp4-tech] D1 @30fps video sequences In-Reply-To: <20050531051814.6584.qmail@webmail52.rediffmail.com> Message-ID: Hi, Can anyone let me know where could I find D1 4:2:0 yuv video sequences? Ideally from surveillance clips. Thanks, Gilberto -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050531/1d129ab3/attachment.html From alexismt comcast.net Tue May 31 19:36:36 2005 From: alexismt comcast.net (Alexis Michael Tourapis) Date: Tue May 31 16:13:25 2005 Subject: [Mp4-tech] A doubt in calculation of PSNR value for Video Message-ID: <053120051836.23092.429CAEB40006D04B00005A3422007621949B039C07980A040E@comcast.net> Dear Ahmad, Your claim is not true. One may for example try to encode a completely black or a completely white image (i.e. consider fades to/from black and white). Other cases could also lead to 0 MSE => infinite PSNR. As Tuukka suggested, one way to avoid problems with the consideration of infinite frame PSNR is to consider the average MSE instead. Other solutions would be to penalize the MSE conditionally (as is done also in the JM) or not (i.e. by always adding a given penalty to the MSE or SSE) in an attempt to avoid infinite values. Obviouslyyou would need to use the same method when comparing two different methods. In general though you would need to be a bit cautious with the consideration of PSNR (or even MSE) as a measure of quality. It is quite possible for example that one may encode a sequence with algorithm/system A and have higher average PSNR than a scheme B, however perceptually B may look better due to fewer temporal or otherwise artifacts imposed by different characteristics of the encoding (normative and/or non normative). Frames that could lead to such "infinite" PSNRs could themselves also artificially spike your average numbers and essentially mask some "bad" PSNR cases that may appear within your sequence (use of average MSE for PSNR computation is more reliable in such cases, i.e. if you wish to experiment, assume you have 3 frames one with MSE equal to 1, a second with MSE equal to 100, and a third with MSE equal to 10. However, note also that averages do not take order of frames in account which could also affect subjective quality). Best regards, Alexis B -------------- Original message -------------- I believe that there is no way that MSE going to be 0. Unless if the compressed frame is identical to original image. Then why do we care calculate PSNR if it is totally the same thing? Am i correct ? From: mp4-tech-bounces@lists.mpegif.org on behalf of M Mahdi Ghandi Sent: Thu 5/26/2005 1:12 AM To: 'Tuukka Toivonen' Cc: mp4-tech@lists.mpegif.org Subject: RE: [Mp4-tech] A doubt in calculation of PSNR value for Video The main problem is when for example this happened for only one frame in a sequence. I will not include that infinite in average PSNR calculation or alternatively replace it with an assumed upper limit (e.g. 50 dB). -----Original Message----- From: mp4-tech-bounces@lists.mpegif.org [mailto:mp4-tech-bounces@lists.mpegif.org] On Behalf Of Tuukka Toivonen Sent: 25 May 2024 08:31 To: T. Bheemarjuna Reddy Cc: mp4-tech@lists.mpegif.org Subject: Re: [Mp4-tech] A doubt in calculation of PSNR value for Video On Tue, 24 May 2005, T. Bheemarjuna Reddy wrote: > I have one basic doubt in calculating PSNR value. > > PSNR (db) = 10 log {(2^n - 1)^2 / MSE} > > what happens to value of PSNR if MSE = 0 ? The PSNR is infinite then. You could check specifically for this case and return the largest possible number. If using IEEE floating points, they can represent infinite exactly. However, normally this is very rare case so people often ignore it. _______________________________________________ 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 The information contained in this Internet message is confidential and intended only for the use of the individual or entity identified. If the reader of this message is not the intended recipient, any dissemination, distribution or copying of the information contained in this Internet message is strictly prohibited. If you received this message in error, please notify the sender immediately. -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/mp4-tech/attachments/20050531/c4b43a05/attachment.html