Methods and devices for geometric partitioning mode with adaptive blending

A method for video encoding, comprising:

• obtaining, by an encoder, a blending index based on a mapping relationship and a blending width of a current block, wherein the current block is partitioned into two parts along a partition line for prediction in a geometric partitioning mode, and the mapping relationship is between a plurality of blending indices and a set of blending widths;
• obtaining, by the encoder, a binarized value by binarizing the blending index, wherein the binarized value comprises one or more bins;
• obtaining, by the encoder, a context compressed binarized value by compressing the binarized value using one or more contexts;
• and
• signaling, by the encoder, the context compressed binarized value in a bitstream.

An apparatus, comprising:

• one or more processors;
• and
• a memory coupled to the one or more processors and configured to store instructions executable by the one or more processors,
• wherein the one or more processors, upon execution of the instructions, are configured to perform operations for video decoding comprising:
• obtaining, by a decoder, a context compressed binarized value generated by compressing a binarized value using one or more contexts, wherein the binarized value is obtained by binarizing a blending index, and comprises one or more bins;
• obtaining, by the decoder, the blending index based on the context compressed binarized value;
• and
• obtaining, by the decoder and based on the blending index and a mapping relationship between a plurality of blending indices and a set of blending widths, a blending width for a current block, wherein the current block is partitioned into two parts along a partition line for prediction in a geometric partitioning mode;
• or
• perform operations for video encoding comprising:
• obtaining, by an encoder, a blending index based on a mapping relationship and a blending width of a current block, wherein the current block is partitioned into two parts along a partition line for prediction in a geometric partitioning mode, and the mapping relationship is between a plurality of blending indices and a set of blending widths;
• obtaining, by the encoder, a binarized value by binarizing the blending index, wherein the binarized value comprises one or more bins;
• obtaining, by the encoder, a context compressed binarized value by compressing the binarized value using one or more contexts;
• and
• signaling, by the encoder, the context compressed binarized value in a bitstream.