Video coding method and system based on neural network

Abstract

The invention discloses a video coding method and system based on a neural network. The method comprises the steps of obtaining video data used for training a neural network, constructing the neural network containing multiple reference frames, training the neural network containing the multiple reference frames based on the video data to obtain a trained neural network, and encoding a video basedon the trained neural network. According to the invention, based on the neural network, the performance of video coding for a low-delay scene can be effectively improved by using multiple reference frames.

Description

technical field [0001] The present invention relates to the technical field of video coding, in particular to a neural network-based video coding method and system. Background technique [0002] Currently, most videos on the Internet are compressed into H.264 or H.265 formats. New video coding standards such as H.266 and AV1 are being developed. Although the new standard saves 50% of the bit rate compared with H.265 under the same visual quality, this performance improvement also brings a multiplied increase in coding complexity. In fact, all traditional coding standards use the same framework. The framework uses motion-compensated prediction, block-based transform, and hand-designed entropy encoder. After more than 30 years of continuous development, while the compression efficiency of the framework has been increasing, the system complexity has become higher and higher, and the performance of individual technologies has become smaller and smaller, which has become satur...

AI Technical Summary

Problems solved by technology

However, these existing models are either oriented to random cut-in scenes, or only use the previous reconstructed frame as a reference frame, and cannot make full use of the temporal correlation between video frames.

When targeting low-latency scenarios, the performance of video encoding is low

Images