Colour component prediction method, encoder, decoder and storage medium

A method for predicting colour components, which is applied to an encoder, the method comprising:

• determining adjacent reference samples of a current block in a picture;
• determining candidate positions of candidate samples based on at least one edge of the current block;
• constructing a subset of adjacent reference samples according to the adjacent reference samples, wherein the subset of adjacent reference samples contains a part of the adjacent reference samples;
• calculating model parameters of a prediction model according to the subset of adjacent reference samples, wherein the prediction model comprises N prediction sub-models, the N prediction sub-models correspond to N groups of model parameters, and the prediction sub-models are used to perform, through corresponding model parameters, cross-component prediction of colour components to be predicted, and N is a positive integer greater than or equal to 2;
• and
• determining a prediction of the current block according to the prediction model;
• wherein determining the adjacent reference samples of the current block in the picture comprises:
• acquiring reference samples adjacent to at least one edge of the current block, wherein the at least one edge of the current block comprises at least one of:
• an upper row, a right upper row, a left column and a left lower column;
• and
• obtaining the adjacent reference samples according to the acquired reference samples;
• wherein constructing the subset of adjacent reference samples according to the adjacent reference samples comprises:
• determining reference samples corresponding to the candidate positions from the adjacent reference samples, and forming the subset of adjacent reference samples with the determined reference samples.