Equipment and methods for encoding or decoding video

Abstract

According to at least one approach, the term "live" refers to the method and equipment for image decoding / prediction, including a prediction step involving at least one block, where the prediction step includes performing motion compensation and local illumination compensation based on a reference block, local illumination compensation includes applying a linear model of the total absolute difference of the reconstructed neighboring samples and the corresponding reference samples of the reference block, In this model, neighboring patterns are reconstructed, and the corresponding reference patterns of the reference block are placed in the same L-shape, essentially adjacent to the predicted block. The L-shape consists of a row of pixels at the top of the predicted block and a column of pixels to the left of the predicted block, with their positions determined by the motion vector of the predicted block.