Issue in pred_mode_ibc_flag parsing

[forayr]

There is a mismatch between the spec and the specification on pred_mode_ibc_flag parsing.
The following modification to the specification are needed to align its behavior with the vtm:

In the Syntax in tabular form:

if( ( ( slice_type == I && cu_skip_flag[ x0 ][ y0 ] == 0 ) ||
 ( slice_type != I && ( CuPredMode[ chType ][ x0 ][ y0 ] != MODE_INTRA ||  
-        ( ( cbWidth == 4 && cbHeight == 4 && cu_skip_flag[ x0 ][ y0 ] == 0 ) ) ) ) &&
+        ( ( ( cbWidth == 4 && cbHeight == 4 ) || treeType == MODE_TYPE_INTRA ) && 
+           cu_skip_flag[ x0 ][ y0 ] == 0 ) ) ) ) &&
    cbWidth <= 64  &&  cbHeight <= 64  &&  modeType != MODE_TYPE_INTER  &&
    sps_ibc_enabled_flag  &&  treeType  !=  DUAL_TREE_CHROMA )
  pred_mode_ibc_flag

In the semantics of pred_mode_ibc_flag:

When pred_mode_ibc_flag is not present, it is inferred as follows:
    • If cu_skip_flag[ x0 ][ y0 ] is equal to 1, and cbWidth is equal to 4, and cbHeight is equal to 4, pred_mode_ibc_flag is inferred to be equal 1.
+   • Otherwise, if cu_skip_flag[ x0 ][ y0 ] is equal to 1, and modeType == MODE_TYPE_INTRA, pred_mode_ibc_flag is inferred to be equal 1.
    • Otherwise, if both cbWidth and cbHeight are equal to 128, pred_mode_ibc_flag is inferred to be equal to 0.
    • Otherwise, if modeType is equal to MODE_TYPE_INTER, pred_mode_ibc_flag is inferred to be equal to 0.
    • Otherwise, if treeType is equal to DUAL_TREE_CHROMA, pred_mode_ibc_flag is inferred to be equal to 0.
    • Otherwise, pred_mode_ibc_flag is infered to be equal to the value of sps_ibc_enabled_flag when decoding an I slice, and 0 when decoding a P or B slice, respectively.

[bbross]

Can someone confirm that aligning the spec to VTM, i.e. adding the condition on modeType introduced by JVET-O0050, is in the spirit of what we adopted w.r.t JVET-O0050?

[zhaoyin]

As the proponent of JVET-O0050, I think the suggested spec changes fix a bug of JVET-O0050 spec.

[bbross]

Thanks for reporting.

This will be fixed JVET-P2001-vD.