VAAC-IM: Viewing Area Adaptive Control in Immersive Media Transmission
Han Xiao, Changqiao Xu, Chuxing Fang, Shujie Yang, Lujie Zhong
Abstract
The viewport in immersive media, exemplified by panoramic video, corresponds to the field of view (FoV) and significantly impacts both the volume of data transmission and the user experience. However, the instantaneous and highly dynamic nature of user interactions poses a challenge to the traditional segment-based transmission mode, creating a conflict between the need for real-time responsiveness and the structured nature of data delivery. It results in the transmission of substantial redundant data, leading to wastage of valuable resources. In this paper, we analyze data from open-source dataset and our self-collected records to investigate the interactive characteristics of viewer in immersive scenes, e.g., focus time, viewing area scope, and movement direction. Based on statistical inferences, we introduce the concept of extended field of view (EoV) to describe irregular viewing areas that mimic the moving features of the human visual field. A motion-aware tile-based viewing area adaptive control scheme, termed VAAC-IM, is designed for transmitting immersive media flexibly. We model viewing area control process as a constrained submodular minimization problem, and design a greedy search-based strategy to dynamically control irregular EoV area. Finally, we perform a comprehensive validation. The results demonstrate that VAAC-IM significantly enhances performance in terms of reducing black edge coverage, minimizing data volume, lowering latency, and improving overall user experience.