Misalignment between Spec and VTM sw on ALF clipping value derivation

[du757372199]

In VVC Spec D6-vE, the ALF clipping values for chroma are derived using the following texts:

alf_chroma_clip_idx[ altIdx ][ j ] specifies the clipping index of the clipping value to use before multiplying by the j-th coefficient of the alternative chroma filter with index altIdx. It is a requirement of bitstream conformance that the values of alf_chroma_clip_idx[ altIdx ][ j ] with altIdx = 0..alf_chroma_num_alt_filters_minus1, j = 0..5 shall be in the range of 0 to 3, inclusive.
The chroma filter clipping values AlfClipC[ adaptation_parameter_set_id ][ altIdx ] with elements AlfClipC[ adaptation_parameter_set_id ][ altIdx ][ j ], with altIdx = 0..alf_chroma_num_alt_filters_minus1, j = 0..5 are derived as specified in Table 7 4 depending on bitDepth set equal to BitDepthC and clipIdx set equal to alf_chroma_clip_idx[ altIdx ][ j ].

However, the Table 7-4 is only used for luma in VTM-6.0, and for chroma, a separate look-up table should be applied as implemented in VTM-6 as well as proposed in JVET-N0242. To align VTM-6 and JVET-N0242, It is suggested to fix the Spec text and use two separate table for deriving the ALF clipping values for luma and chroma.

[patrice.onno]

I think that it is the other way around.

Per the meeting notes of the 15th JVET meeting, a decision has been taken to have a single table of clipping values for both Chroma and Luma components. This table is listing the 4 clipping values for the different bitdepths. The VVC specifications text and the corresponding Table 7-4 in the VVC draft 6 (JVET-O2001, version 14) is in agreement with this decision.

However, the VTM6.0 software has not been updated to reflect this change.

We can provide a patch for that.

[bbross]

This seems to be a VTM issue which might be already solved.

[fbossen]

This doesn't seem relevant anymore.