[Mp4-tech] the CAVLC code design in MPEG4 (part10)

[yux porchlight.ca]

Thanks a lot, Richardson,
I have read your tutorial. It is very helpful. The image of
what to do to implement the code is clear. I am wondering
if you could help me with some further questions, as how the
code is designed as like below. Thanks in advance.
1. How is the context represented while designing the lookup tables?
  I may guess that some empirical distributions are built up
  for non-zero coefficient in all positions? E.g., p(x1) for
  the first non-zero quantized coefficient, p(x2) for the second.
  (The first here is the last one in the zigzag as the coding
  process is carried on in the inverse order).
  Is this right for the CAVLC design in MPEG4(part10)?
  If yes, is it possible for me to find a report on this?
  If no, how is the context represented and formed?
2. What is the design criterion for these tables? For example,
  what is the point behind the '0...01x...xs' structure of Level
  VLC1-6? Or, what is done from the context to code tables?
3. What is the criterion to determine the table switch threshold
  for VLC0-6? Or how does it depend on the number of non-zero
  coefficients in neighbouring blocks?
Thanks again and best regards,
shaun
> You might find my tutorial useful:
> www.vcodex.com/h264.html
>
> The terminology for describing CAVLC is slightly different in the latest
> version of the standard but the outcome is the same.
>
> Regards
>
> Iain Richardson
>
> -----Original Message-----
> From: mp4-tech-bounces lists.mpegif.org
> [mailto:mp4-tech-bounces lists.mpegif.org] On Behalf Of yux porchlight.ca
> Sent: 24 November 2023 21:18
> To: Mp4-tech lists.mpegif.org
> Subject: [Mp4-tech] the CAVLC code design in MPEG4 (part10)
>
>
> Hi, there,
>
> Would someone on the list give some information on how the CAVLC code is
> designed in MPEG4, part 10/H.264, (how and what kind of prior knowledge is
> used to build up those VLC tables for the residual coding)? I am studying
> lossless source coding techniques and would like to study the informatic
> redundancy rate for this code.
>
> Any informationa would be highly appreciated!
>
> Shaun
> _______________________________________________
> Mp4-tech mailing list
> Mp4-tech lists.mpegif.org
> http://lists.mpegif.org/mailman/listinfo/mp4-tech
>
> ---
> Incoming mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.537 / Virus Database: 332 - Release Date: 06/11/2023
>
>
> ---
> Outgoing mail is certified Virus Free.
> Checked by AVG anti-virus system (http://www.grisoft.com).
> Version: 6.0.537 / Virus Database: 332 - Release Date: 06/11/2023
>
>
>