360-degree video stream transmission system based on saliency detection

Abstract

The invention discloses a 360-degree video stream transmission system based on significance detection, and the system comprises a server end which obtains a significance matrix according to eye movement data at an offline stage of the system and calculates the variance of each video, i.e., the average variance of the significance matrix, thereby enabling the videos to be divided into a significance video and a non-significance video; after acquisition and analysis of the significance matrix are completed, the video exists in different qualities and corresponding forms through fragmentation inthe time dimension and the space dimension; and a client which is used for downloading the corresponding saliency matrix from the server to the client player according to the actual network conditionwhen a user watches the saliency video, controlling the downloading and transmission of the video stream in the playing process based on the saliency code rate control algorithm, and adjusting the quality of the video stream through the MPC algorithm for the non-saliency video. According to the method, the defects that the cache length is too small due to nonlinearity of user watching behaviors and limitation of a prediction window and the like are effectively relieved.

Description

technical field [0001] The invention belongs to the field of network systems, and in particular relates to a 360° video stream transmission system based on saliency detection. Background technique [0002] With the increasing popularity of 360° videos, virtual reality (Virtual Reality, VR) and augmented reality (Augmented Reality, AR) technologies have begun to receive attention from all walks of life. In recent years, with the maturity of VR equipment, network and terminal technologies, 360° videos are increasingly used in applications such as film and television, games, medical care, and education. According to Cisco's statistical report, the average monthly traffic of VR applications reached 13.3PB in 2016, and is expected to reach 140PB by 2021, with an annual growth rate of 60%. Such a wide range of 360° videos has brought new experiences to people's life and work, and a large number of network infrastructure and smart wearable devices have also promoted the developmen...

AI Technical Summary

Problems solved by technology

However, the 360° video stream transmission scheme based on head movement trajectory prediction has certain disadvantages due to the nonlinearity of user viewing behavior and the limitation of the prediction window, which makes the buffer length too small.

Images