FIELD: video technology.
SUBSTANCE: invention relates to the field of encoding/decoding, in particular to encoding in the intra-mode of the High-Efficiency Video Signal Encoding (HEVC) standard. The output of the reference values of the prediction mode, also called the most probable modes (MPM), which are used for encoding/decoding the prediction mode associated with the current encoding element, is proposed. The first and second reference values of the prediction mode are derived from the corresponding prediction modes of at least two neighboring encoding elements of the current encoding element. The first and second reference modes of prediction differ from each other, and the third reference value of the prediction mode is derived from the first and second reference values of the prediction mode. In this case, the first and third reference values of the internal prediction mode are used to encode or decode the value of the mode to be encoded or decoded. By outputting three MPMs instead of two for comparison with the prediction mode of the current encoding block, the encoding efficiency is improved, since there is an increase in the probability that the prediction mode of the current encoding block corresponds to one of the most probable modes derived.
EFFECT: increased efficiency of encoding/decoding information about prediction modes.
25 cl, 11 dwg
Authors
Dates
2021-11-11—Published
2020-12-08—Filed