MADRL-Based Rate Adaptation for 360{\deg} Video Streaming with Multi-Viewpoint Prediction

TL;DR

This paper introduces a multi-viewpoint prediction model using a multimodal transformer and a MADRL-based adaptive streaming algorithm, significantly enhancing 360-degree video QoE under network constraints.

Contribution

It proposes a novel multi-viewpoint prediction approach with attention mechanisms and a MADRL-based ABR algorithm for improved 360° video streaming.

Findings

QoE improved by up to 85.5% over existing methods

Multi-viewpoint prediction effectively captures user head movement

The approach adapts well to various network conditions

Abstract

Over the last few years, 360{\deg} video traffic on the network has grown significantly. A key challenge of 360{\deg} video playback is ensuring a high quality of experience (QoE) with limited network bandwidth. Currently, most studies focus on tile-based adaptive bitrate (ABR) streaming based on single viewport prediction to reduce bandwidth consumption. However, the performance of models for single-viewpoint prediction is severely limited by the inherent uncertainty in head movement, which can not cope with the sudden movement of users very well. This paper first presents a multimodal spatial-temporal attention transformer to generate multiple viewpoint trajectories with their probabilities given a historical trajectory. The proposed method models viewpoint prediction as a classification problem and uses attention mechanisms to capture the spatial and temporal characteristics of input video frames and viewpoint trajectories for multi-viewpoint prediction. After that, a multi-agent deep reinforcement learning (MADRL)-based ABR algorithm utilizing multi-viewpoint prediction for 360{\deg} video streaming is proposed for maximizing different QoE objectives under various network conditions. We formulate the ABR problem as a decentralized partially observable Markov decision process (Dec-POMDP) problem and present a MAPPO algorithm based on centralized training and decentralized execution (CTDE) framework to solve the problem. The experimental results show that our proposed method improves the defined QoE metric by up to 85.5% compared to existing ABR methods.