A client-based adaptation framework for 360-degree video streaming

Abstract 360-degree video is one of the key components of Virtual Reality (VR) applications. 360-degree videos viewed on Head Mounted Displays can offer impressive viewing experiences to users. Yet, streaming of 360-degree videos over the Internet is a very challenging task since it requires extremely high bandwidth. To reduce the bandwidth requirement while still providing good experiences, viewport adaptive streaming has been introduced. In this paper, we propose a client-based adaptation framework for viewport adaptive streaming of 360 videos, which can support different application scenarios. The key components and important issues are presented with a general problem formulation for tile version selection. Especially, we introduce for the first time the use of bitrate and quality estimation in viewport adaptive streaming of 360 videos. Experiments show that the proposed framework can significantly improve video quality for users. In addition, the impacts of buffering delay and projection formats in VR context are investigated.