Image filtering, encoding and decoding method and related equipment

Abstract

The invention discloses an image filtering, encoding and decoding method and related equipment. Comprising the following steps: performing multi-stage down-sampling on a reconstructed block to obtain a thumbnail feature of the reconstructed block; the main features and the auxiliary features which are the same in size are subjected to down-sampling in each stage of down-sampling, and the sampled main features are fusion features of the main features and the auxiliary features obtained through down-sampling in the previous stage; performing multi-stage up-sampling on the thumbnail features to obtain a filtering block of the reconstruction block; the main features and the auxiliary features which are the same in size are subjected to up-sampling in each level of up-sampling, the sampled main features are fusion features of the main features and the auxiliary features obtained through upper-level up-sampling and the down-sampling main features of the corresponding size, and the filtering block is the main features obtained through last-level up-sampling. Through the above mode, the feature extraction capability can be optimized, and the filtering effect can be improved.

Description

technical field [0001] The present invention relates to the technical field of video encoding, in particular to an image filtering method, an encoding method, a decoding method, an encoder, a decoder, and a computer-readable storage medium. Background technique [0002] The amount of video image data is relatively large. Usually, the video pixel data needs to be compressed to form a video code stream. The video code stream is transmitted to the client through a wired or wireless network, and then decoded for the user to watch. The entire video encoding process includes block division, prediction, transformation, quantization, encoding and other processes, and the video pixel data can also be filtered to make the image more natural. However, the existing image filtering methods are relatively simple. Usually, the image input is simply convolved, activated, etc., and the filtering effect is poor, and there is still a large room for improvement in feature extraction capabilitie...

AI Technical Summary

Problems solved by technology

However, the existing image filtering methods are relatively simple. Usually, the image input is simply convolved, activated, etc., and the filtering effect is poor, and there is still a lot of room for improvement in feature extraction capabilities.

Images