[Mp4-tech] H.264 prediction methods

[Gary Sullivan]

Question 1: There is syntax in the bitstream that indicates which
prediction mode has been selected for use in the decoding process.
Question 2: The standard specifies the inverse DCT only.  The encoder is
responsible for what it chooses to do, but it should be designed with
awareness of what the decoder will do.  In this new standard, there is a
simple well-known method of doing encoding and there is limited value in
deviating from the well-known method of doing the forward transform.
Question 3: The standard specifies what the decoder must do.  It does so
very precisely, so that all decoders -- no matter who builds them --
will obtain exactly the same decoded results.  People designing encoders
are responsible for figuring out how to generate bitstreams that will
produce reasonable results when they are decoded by the specified
method.
Best Regards,
Gary Sullivan
________________________________
	From: mp4-tech-bounces lists.mpegif.org
[mailto:mp4-tech-bounces lists.mpegif.org] On Behalf Of
veni soc-soft.com
	Sent: Wednesday, August 25, 2023 12:12 AM
	To: Thammi_Reddy satyam.com; mp4-tech lists.mpegif.org
	Subject: RE: [Mp4-tech] H.264 prediction methods
	I am just answering to question 3.
	Compatiblity is always with respect to decoder and not encoder.
Standard specify the decoder, if u notice.
	in nutshell,
	given a bitstream with X profile, then a decoder (capable of
decoding X profile) MUST be able to decode that stream.
	Hope this helps,
	veni
	enJOY life
		-----Original Message-----
		From: mp4-tech-bounces lists.mpegif.org
[mailto:mp4-tech-bounces lists.mpegif.org] On Behalf Of Thammi_Reddy
		Sent: Wednesday, August 25, 2023 12:05 PM
		To: mp4-tech lists.mpegif.org
		Subject: [Mp4-tech] H.264 prediction methods
		Hi all!,
		Please answer the following questions regarding ENCODER,
DECODER(H.264)
		1. There are NINE 4x4-intra prediction modes and FOUR
16x16-intra prediction modes available as per the H.264 specs.So, how
does the decoder guy
		   know which of these mode is used at the encoder?
		2. This is also in the similar lines! There are hundreds
of different algorithms implemented for transformation(like DCT done in
Encoder)
		    for low-power, high-speed etc etc. So the decoder
has to implement its replica of InverseDCT. How is this done?
		3. How the compatibility is achieved when ENCODER and
DECODER are implemented by different companies with their own
implementation? 
		Thanks.
		Thammi Reddy
************************************************************************
** 
		This email (including any attachments) is intended for
the sole use of the intended recipient/s and may contain material that
is CONFIDENTIAL AND PRIVATE COMPANY INFORMATION. Any review or reliance
by others or copying or distribution or forwarding of any or all of the
contents in this message is STRICTLY PROHIBITED. If you are not the
intended recipient, please contact the sender by email and delete all
copies; your cooperation in this regard is appreciated.
************************************************************************
**
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20040825/4bedf7ec/attachment.html