Encoding method, decoding method, code stream, encoder, decoder and storage medium

A decoding method, applied to a decoder, the method comprising:

• parsing a bitstream, and determining a value of first syntax element identification information;
• in response to the first syntax element identification information indicating that model selection using a preset selection network model is enabled for a current block, determining the preset selection network model for the current block, determining an output value corresponding to each of at least one candidate in-loop filtering network model according to the preset selection network model, and determining an in-loop filtering network model used for the current block according to the output value corresponding to each of the at least one candidate in-loop filtering network model;
• and
• acquiring a reconstructed picture block of the current block through filtering the current block by using the in-loop filtering network model,
• wherein the preset selection network model consists of a plurality of convolutional neural network (CNN) layers and a plurality of fully connected neural network layers.