JVET-P1019: Output layer set

Brian Heng authored 5 years ago and Karsten Suehring committed 5 years ago

  - Remove bricks and related syntax.
  - Add new tile / slice syntax from P1004.
  - Fix cross tile / slice prediction bugs.
  - Update config files and documentation with new tile / slice config options.

Brian Heng authored 5 years ago and Karsten Suehring committed 5 years ago

  - Add picture header and related syntax changes.

1. enabling the encoder-side RD check of ISP mode when ACT is disabled at CU level
2. ACT intra RD code clean-up

Detail
- Fix the mis-alignment of rate and distortion during palette coding RD decision.
- Use different weights on luma component cost and chroma component cost.

I have confirmed that this modification can be consistent with the specification and there is no problem with the coding results.

  - Align MLRP with CCLM in terms of reference lines/columns.

Karsten Suehring authored 5 years ago and Jizheng Xu committed 5 years ago

- create ceilLog2() function based on floorLog2
- use intrinsics for MSVC in floorLog2
- replace all ceil(log2()) calls in HLS with ceilLog2()

- create ceilLog2() function based on floorLog2
- use intrinsics for MSVC in floorLog2
- replace all ceil(log2()) calls in HLS with ceilLog2()

Remy Foray authored 5 years ago and Frank Bossen committed 5 years ago

There is no scaling matrix for palette prediction

Software changes with respect to the adoption from proposal JVET-O0455:
Control the maximum number of IBC merge candidates independently from
the maximum number of regular merge candidates.