Adaptive code rate video transmission method and system based on reinforced learning

Abstract

The invention discloses an adaptive code rate video transmission method and an adaptive code rate video transmission system based on reinforced learning. The method comprises the steps of inputting astate space corresponding to a video block needing to be downloaded into a code rate prediction neural network, and outputting a code rate strategy by the code rate prediction neural network; downloading the video block needing to be downloaded according to the code rate strategy output by the code rate prediction neural network; after each video block is completely downloaded, computing a video playing quality index corresponding to each video block and returning the video playing quality index back to the code rate prediction neural network; and performing training by the code rate prediction neural network according to the returned video playing quality index and the state space corresponding to the completely downloaded video block. According to the method and the system provided by the invention, the algorithm well adapts to multiple different network conditions while the quality of service is improved, the labor time cost of rule setting and parameter optimization is greatly reduced, the efficiency problem caused by artificial feature selection and rule setting is avoided, and the video quality experience is also greatly improved while the configuration and debug time is saved.

Description

technical field [0001] The present invention relates to the field of video transmission, in particular to a reinforcement learning-based adaptive code rate video transmission method and system. Background technique [0002] With the development of wireless networks in recent years, video applications based on the HTTP protocol have become more and more abundant. What follows is that users' demand for video quality is also constantly increasing. If the video quality is not good, users will quickly give up watching. Therefore, how to provide high-quality videos to users has become a key issue for operators of major video platforms. one of the research fields. [0003] Adaptive bit rate algorithm is the main method used by content providers to optimize video quality. The existing mainstream adaptive bit rate algorithm selects bit rate by estimating network throughput and combining control rules, which requires a lot of tuning work and is difficult Generalization under differe...

AI Technical Summary

Problems solved by technology

[0003] Adaptive bit rate algorithm is the main method used by content providers to optimize video quality. The existing mainstream adaptive bit rate algorithm selects bit rate by estimating network throughput and combining control rules, which requires a lot of tuning work and is difficult Generalization under different network and quality of service objectives

Images