AOMedia's "AVM" Repository Serves As Reference Implementation For Eventual AV1 Successor

[phoronix]

Phoronix: AOMedia's "AVM" Repository Serves As Reference Implementation For Eventual AV1 Successor

The Alliance for Open Media already has a public repository serving as the reference software implementation for "AVM" as the AOM Video Model as the next-generation codec from this alliance responsible for the excellent AV1...

https://www.phoronix.com/scan.php?page=news_item&px=AOMedia-AVM-Next-Gen-Repo

[cl333r]

I imagine the biggest issue is to develop a next-gen solution while not stepping on any landmine in the giant and complicated patent minefield.

[brucethemoose]

IIRC there was talk of incorporating neural nets into parts of the next gen codec.


I know AI gets thrown around as a buzzword and slapped on things that don't deserve the term, but there are stages of the AV1 pipeline that it could really dramatically help.

[CommunityMember]

I imagine the biggest issue is to develop a next-gen solution while not stepping on any landmine in the giant and complicated patent minefield.

May (or may not) be successful in avoiding the minefield (as you say, it is complicated). AV1 already has an organization (Sisvel) offering a patent pool license which they claim is required to use AV1. The legal team for the Alliance for Open Media has TTBOMK not (yet) responded to the patent claims.

[wswartzendruber]

May (or may not) be successful in avoiding the minefield (as you say, it is complicated). AV1 already has an organization (Sisvel) offering a patent pool license which they claim is required to use AV1. The legal team for the Alliance for Open Media has TTBOMK not (yet) responded to the patent claims.

Ah yes, Sisvel. The entity that said they'd disclose all patents needed for AV1 and then didn't do anything.

[direc85]

How about fixing and completing the acceleration of current-gen video codecs on current-gen hardware first? What use is a new shiny codec if I can't use Firefox with VA-API at all?

[wswartzendruber]

I just want the damned thing (SVT-AV1, aomenc, WHATEVER!!!) to handle noise modeling correctly.

SVT-AVT1 is almost there. I'm tempted to fork and do some code diving to see if I stand any chance at fixing it.

[bug77]

How about fixing and completing the acceleration of current-gen video codecs on current-gen hardware first? What use is a new shiny codec if I can't use Firefox with VA-API at all?

This is pipelined work. It doesn't mean the current version gets no attention anymore, if someone started thinking about laying the ground work for the next generation.

[Mathias]

IIRC there was talk of incorporating neural nets into parts of the next gen codec.

I agree that some next generation video codec will have an AI enhancement layer baked in. I guess this will improve the quality for very low bitrates significantly. IMO one of the biggest issues will be what metric to use to judge quality improvements. AI can generate good looking pictures that don't resemble the original well, how will the codec and the developers know something looks good, if it doesn't have to resemble the original perfectly? We need an AI based metric for that as well. Which then has the problem that an AI judges what an AI generates. So that metric has to be exceptionally good. So that needs a lot of training data...

That will of course also make comparisons to other codecs difficult. Because if one codec is tuned for a specific metric, it of course beats every other codec tuned to a different metric. This alone will give "a 300% improvement in image quality!!!!".

[abu_shawarib]

How about fixing and completing the acceleration of current-gen video codecs on current-gen hardware first? What use is a new shiny codec if I can't use Firefox with VA-API at all?

Developing such codecs is a long process. If you wait until the previous spec is completely adopted before starting the work on the next one you'll be 5 years behind the competition. Also, quite a few of the members of AOM don't make decoding/encoding hardware.