Metadata-Based Quality-Estimation Model for Tile-Based Omnidirectional Video Streaming

In this paper, a metadata-based quality-estimation model is proposed for tile-based omnidirectional video streaming services, aiming to realize quality monitoring during service provision. the (ODV) ODV divided into tiles, and high-quality tiles low-quality are distributed in accordance with user's viewing direction. When user changes direction, temporarily watches tiles. addition, longer time (delay time) until tile new direction downloaded, of tile, thus delay affects quality. From above, significantly impact quality, these factors need be considered model. We develop models by extending conventional 2D adaptive streaming. also show that using bitrate, resolution, frame rate high- has sufficient estimation accuracy based on results subjective evaluation experiments.