International organisation for standardisation organisation internationale de normalisation

5.17.2Intra prediction

5.17.2.1Chroma intra prediction

5.17.2.1.1.1.1.1.1JCTVC-H0177 Non-CE6a: Use of chroma phase in LM mode [E. François, S. Pautet, C. Gisquet (Canon)]

This proposal provided additional results related the chroma phase issue in the luma-based chroma prediction mode (LM mode), already addressed in contribution JCTVC-G245. The issue consists in a possible misalignment of interpolated luma samples with the chroma samples. In JCTVC-G245, it has been proposed to signal the chroma phase parameters and to adapt the luma interpolation process based on this chroma phase information.

This proposal reports further experiments and recommends a set of filters to handle four possible basic chroma phase settings. The evaluation has been made on different versions of 4:2:0 sequences, generated from 4:4:4 sequences, whose chroma components have been downsampled with specific chroma phases. Using these recommended filters, average BD BR gains up to* 0.2% Y, 4.2% U and 3.5% V are reported.

(* = the set of 4:4:4 sequences have been converted into several sets of 4:2:0 sequences with different chroma phases. One average BD BR gain is measured for each set of 4:2:0 sequences.)

It was asked whether this is practical, and how an encoder could determine the correct chroma phase.

Concern was expressed that this could make decoders more complex. It appeared that it could be relevant for professional applications. However, no action was taken on this.

5.17.2.1.1.1.1.1.2JCTVC-H0475 Signalling chroma intra prediction mode [J. Kim, B. Jeon (LG)]

Discussed in BoG.

Decision: Adopt binarization scheme for chroma prediction mode from H0475 (the same as was proposed as part of H0326).

5.17.2.1.1.1.1.1.3JCTVC-H0339 Non-CE6: Crosscheck for LG intra chroma mode codeword design in JCTVC-H0475 [T.-D. Chuang, Y.-W. Huang (MediaTek)] [late]

5.17.2.1.1.1.1.1.4JCTVC-H0691 non-CE6.a: A joint contribution of cross-channel prediction method for chroma [K. Kawamura, T. Yoshino, H. Kato, S. Naito (KDDI), Y. Chiu, Y. Han, L. Xu, W. Zhang, H. Jiang (Intel)] [late]

This combines H0117 and H0295, no further review was necessary, as H0295 was not considered.