Scalable Video Coding Based on User’s View for Real-Time Virtual Reality Applications

Abstract

The transmission of virtual reality (VR) videos requires huge bandwidth, which brings great challenges for the system to implement real-time applications. This letter proposes a scalable full-panorama video coding method to adapt the insufficient bandwidth, in which user's movement information is utilized as a feedback from the VR device to the video encoder. The regions that the user is interested in are first mapped and then coded in high quality, while the others in low quality. This different-quality coding is achieved through scalable high efficiency video coding (SHVC), which only makes limited modifications on the SHVC encoder and no modification on the SHVC decoder. In addition, full-panorama coding also avoids the reference problem in inter prediction and relieves the delay-sensitive problem. Experiments results show that our proposed method reduces approximately 87% bit rate with no significant decrease in quality of the region of interest in panorama.