[Help Wanted] Add support for dynamic frame pyramid sizes #3158
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Currently rav1e (assuming no low latency mode) will always code a mini-GOP with 4 input frames with a constant pyramid structure. This changeset adds support to rav1e for variable width and depth frame pyramids, whose size is selected based on frame heuristics; i.e. segments with less motion may have both wider and taller pyramids without losing quality, segments with more motion must have shallower pyramids to preserve quality.
Currently the framework for this changeset is setup, but I have been stuck for some time on figuring out the proper way to code the reference frame values, see
src/encoder.rs:1040-1080
. Any amount of help on this from rav1e knowers would be appreciated.