[x264-devel] Regarding frame boundaries in streaming applications.

[Loïc Le Loarer]

Le jeudi 05 juillet 2007 à 09:35:35 +0530, mahesh k a écrit:
> Hi all,
> 
> Can any one plz tell me whether the syntax specified in Annex B from
> standard is suitable for streaming applications.

Of course, not having this feature would render H.264 unusable for many
applications like IPTV which it has been designed for.

> Also let me know how to identify the frame boundaries in streaming
> applications (in the sense that there is no need to give the encoded frame
> lengths as input to the decoder.

Read chapter "7.4.1.2.4 Detection of the first VCL NAL unit of a primary
coded picture", and Annex B. Read also "D.2.7 Recovery point SEI message
semantics" because with IDR, recovery point are the only official entry
points in a H.264 stream (an entry point is a frame at which you can
start decoding correctly the stream).

> Also let me know the difference between end of seuence NAL unit and  end
> of stream NAL unit.

As its name indicate, "end of sequence" is only the end of a sequence,
which can be followed by a new sequence in the same stream. While end of
file can occur only once in a file.

There is much information on the web about H.264 and video decoding in
general which can answer this kind on questions. You should read at
least the wikipedia H.264 entry http://en.wikipedia.org/wiki/H.264.
You could also read the Annexe D of MPEG-1 norm
http://www.csie.ntu.edu.tw/~b90013/ejection/course/4down/11172-2.pdf
it is very informative about general video encoding/decoding.

Best regards.

-- 
Loïc

"heaven is not a place, it's a feeling"
_______________________________________________
x264-devel mailing list
x264-devel at videolan.org
http://mailman.videolan.org/listinfo/x264-devel