Vues

• DOI: 10.1145/3495243.3517027
• Publication Date: 2022-10-14T15:38:33Z
• Authors: Yu Liu; Bo Han; Feng Qian; Arvind Narayanan; Zhi-Li Zhang
• Keywords: Transcoding; Frame rate

The emerging volumetric videos offer a fully immersive, six degrees of freedom (6DoF) viewing experience, at the cost extremely high bandwidth demand. In this paper, we design, implement, and evaluate Vues, an edge-assisted transcoding system that delivers high-quality with low requirement, decoding overhead, quality experience (QoE) on mobile devices. Through IRB-approved user study, build first-of-its-kind QoE model to quantify impact various factors introduced by content into 2D videos. Motivated key observations from Vues employs novel multiview approach overarching goal boosting QoE. edge server adaptively transcodes video frame multiple views help few lightweight machine learning models strategically balances extra consumption additional improved QoE, indicated our model. client selects view optimizes among delivered candidates for display. Comprehensive evaluations using prototype implementation indicate dramatically outperforms existing approaches. On average, it improves 35% (up 85%), compared single-view schemes, reduces 95%, state-of-the-art directly streams