JVET-R0066: Signal ols_dpb_chroma_format[ i ] and ols_dpb_bitdepth_minus8[ i ] in the VPS,...

1. Signal the maximum values of chroma_format_idc and bit_depth_minus8 for all pictures of all layers in the VPS
2. When sps_video_parameter_set_id is greater than 0, it is required that, for any OLS with OLS index i that contains one or more layers that refers to the SPS, the value of chroma_format_idc shall be less than or equal to the value of ols_dpb_chroma_format[ i ], and the value of bit_depth_minus8 shall be less than or equal to the value of ols_dpb_bitdepth_minus8[ i ].
3. The value of no_output_of_prior_pics_flag, when present, is required to be the same for all pictures in an AU

changed the description

added 27 commits

• b30f9a92...e47497e3 - 24 commits from branch jvet:master
• 78ba51b6 - signal ols_dpb_chroma_format[ i ] and ols_dpb_bitdepth_minus8[ i ] in the VPS,...
• 6d31974d - rebase
• 9c77b300 - Merge branch 'R0066' of https://vcgit.hhi.fraunhofer.de/ZhipinDeng/VVCSoftware_VTM into R0066

Compare with previous version

added 48 commits

• 9c77b300...5dda2bc7 - 43 commits from branch jvet:master
• 18688f16 - signal ols_dpb_chroma_format[ i ] and ols_dpb_bitdepth_minus8[ i ] in the VPS,...
• ad855b46 - rebase
• 437f771b - signal ols_dpb_chroma_format[ i ] and ols_dpb_bitdepth_minus8[ i ] in the VPS,...
• 0e9c392d - rebase
• 494b69e3 - commit

Compare with previous version

merged

mentioned in commit 933b9936

Is it necessary to add a new entry for every slice in a picture? Doing so seems to cause issues when the value of NoOutputOfPriorPicsFlag is modified between slices during "the inference for NoOutputOfPriorPicsFlag" later in the xDecodeSlice function, causing the decoder to crash for multi-slice streams.

There is a mistake in the code. I pushed MR !1696 (merged) to fix it. Would you please have a check? Thanks.

mentioned in merge request !1696 (merged)

mentioned in commit cb7b243e