Head Movement-aware MPEG-DASH SRD-based 360° Video VR Streaming System over Wireless Network
PubDate: August 2022
Teams: Pohang University of Science and Technology；Dongguk University
Writers: Seunggyu Ji; Seunghwan Lee; Giseok Park; Hwangjun Song
This paper presents an effective adaptive 360° video VR streaming system based on the Spatial Representation Description (SRD) feature of MPEG-DASH. The goal of the proposed system is to provide a seamless 360° video VR streaming service with high spatial quality over the time-varying wireless network. To achieve this goal, the proposed system adopts tile-based viewport-adaptive streaming that is a promising way to efficiently deliver 360° video. In accordance with the SRD information and user’s head movement, the proposed system is designed to enhance the quality of the user experience by intensively allocating a limited bit budget to specific segments of interest without noticeable blanks. Additionally, we propose a fast recovery algorithm to minimize the effect of blanks caused by unpredictable head movement. Moreover, the proposed system considers the binocular disparity of the human eye to improve the subjective quality of stereo video. The proposed system is fully implemented by using well-known open sources (e.g., OpenHMD, Kvazaar and GPAC), C/C++, Python and head-mounted display (HMD) equipment. Finally, the experimental results are provided to verify the performance of the proposed system in real wireless network environments.