RAPT360: Reinforcement Learning-Based Rate Adaptation for 360-Degree Video Streaming With Adaptive Prediction and Tiling

• Published in: IEEE transactions on circuits and systems for video technology Vol. 32; no. 3; pp. 1607 - 1623
• Main Authors: Kan, Nuowen, Zou, Junni, Li, Chenglin, Dai, Wenrui, Xiong, Hongkai
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.03.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: 360-degree videos; Adaptation; Adaptation models; Adaptive systems; adaptive viewport prediction; Algorithms; Bandwidth; Bit rate; Confidence intervals; Deep learning; deep reinforcement learning; Forecasting; Identification methods; Machine learning; Markov processes; Occupancy; Optimization; Prediction algorithms; Probability density functions; Quality of experience; rate adaptation; Statistical analysis; Streaming media; tile-based adaptive streaming; Tiles; Tiling; Two dimensional models; Video transmission; viewport-aware tiling

Tile-based rate adaption can improve the quality of experience (QoE) for adaptive 360-degree video streaming under constrained network conditions, which, however, is a challenging problem due to the requirements of accurate prediction for users' viewports and optimal bitrate allocation for tile...