Video quality evaluation method and system, computer equipment and storage medium

Abstract

The invention provides a video quality evaluation method and system, computer equipment and a storage medium, and the method comprises the steps: obtaining a video data set, and randomly dividing the video data set into a training set and a test set according to a preset proportion; inputting the training set into a preset video quality evaluation network for training to obtain a video quality evaluation model, the video quality evaluation network sequentially comprising a preprocessing module, an attention coding module and a vector regression module, smf the attention coding module comprising a plurality of space-time attention coding modules; and predicting the test set by using the video quality evaluation model to obtain a corresponding predicted video quality score. According to the method, the quality evaluation problem of the naturally distorted video is effectively solved, the long-range space-time dependence characteristic in the video sequence is efficiently captured, the training convergence speed and the video evaluation performance are improved, and the application field of the Transform architecture is expanded.

Description

technical field [0001] The invention relates to the technical fields of video processing and computer vision, in particular to a video quality evaluation method, system, computer equipment and storage medium based on a space-time attention mechanism. Background technique [0002] With the vigorous development of social media, video content captured by mobile users has exploded on major social platforms, such as TikTok, Facebook, Instagram, YouTube, and Twitter. The storage, transmission and processing of these massive contents have brought great challenges to the video service industry, especially the influx of low-quality videos shot by amateur users in harsh environments to the Internet, which not only affects people’s video perception, but also affects people’s ability to access Useful information in the video. Therefore, using automated video quality assessment tools to filter low-quality videos and prevent them from spreading and affecting video service effects has gra...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a video quality assessment method, system, computer equipment and storage medium based on a space-time attention mechanism, and to establish video quality by appropriately improving the pure Transformer architecture that does not contain any convolution operations based on video assessment requirements Evaluate the network (StarVQA network), and use the StarVQA network to capture the long-term space-time dependence from the input video frame, so as to effectively solve the natural distortion video quality evaluation problem and improve the video evaluation performance

Images