Cross-component prediction in multi-partition prediction modes

Abstract

The various embodiments described in this application include methods and systems for video encoding and decoding. In one aspect, the video bitstream includes a current coded block of a current image frame and a first syntax element for a multi-segment prediction mode. Based on the first syntax element, a computational system determines that a multi-segment prediction mode is enabled to reconstruct the current coded block based on multiple segments. Each of the multiple segments corresponds to a set of corresponding model parameters used to reconstruct a corresponding chroma sample based on a set of corresponding luminance samples. The computational system determines that a first chroma sample is located in a first segment, combines a set of first luminance samples using a set of first model parameters to generate a first chroma sample, and reconstructs the current image frame including the first chroma sample of the current coded block.

Images