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

There is no scaling matrix for palette prediction

fix compilation errors by code merging and harmonize with JVET_O0119_BASE_PALETTE_444, JVET_O0502_ISP_CLEANUP, JVET_O0472_LFNST_SIGNALLING_LAST_SCAN_POS, JVET_O0545_MAX_TB_SIGNALLING

- only cu_qp_delta is conditioned by (CbWidth > 64 || CbHeight > 64)
- move parsing of tu_joint_cbcr_residual_flag[] between cu_chroma_qp_offset_idx and transform_skip_flag instead of before cu_qp_delta

  - Add back support for reference wraparound with DMVR.

fix some JCCR related indentation, encapsulation, and descriptive text, add --PYUV to software manual

- refactoring of MatrixIntraPrediction.h/.cpp (merged classes MatrixIntraPrediction and PredictorMIP, removed namespace Mip)
- removed unused variable posBias in function computeReducedPred
- fixed list size in function reduceHadCandList

This reverts commit 9503ca9c

  - Prevent weighted prediction related processing when IBC or triangle prediction is enabled for the PU.

- fixed build error

Fix #426: Chroma QP mapping tables - incorrectly initialized arrays for some encoder configurations.

- Fix for the case that uninitialized or invalid best cost values are referred during search candidate selection

numCandminus1_step is always positive

  - Reduce DMVR SAD cost by 2x to match spec.
  - Adjust BDOF-Disble and Zero-Cost thresholds accordingly.

This part was originally integrated in HEVC SCM8.0 but was later removed as a bugfix in SCM8.3.....

Encoder side bug fix: the encoder speed up in palette (which bypass INTRA mode check in inter slice if palette mode is better than INTER mode)
shouldn't be applied on 4x4 blocks since JVET-N0266 disabled INTER mode in 4x4 blocks.
Without the fix, intra mode will always be skipped in 4x4 blocks in inter slice if IBC is disabled in SPS.

BugFix of encoder deblocking optimization: add initialization of m_ctuXLumaSamples and m_ctuYLumaSamples, align with 4x4 deblocking

This reverts commit 9503ca9c