Adaptive 360 VR Video Streaming based on MPEG-DASH SRD

Mohammad Hosseini,Viswanathan Swaminathan

DOI: https://doi.org/10.48550/arXiv.1701.06509

2017-01-24

Abstract:We demonstrate an adaptive bandwidth-efficient 360 VR video streaming system based on MPEG-DASH SRD. We extend MPEG-DASH SRD to the 3D space of 360 VR videos, and showcase a dynamic view-aware adaptation technique to tackle the high bandwidth demands of streaming 360 VR videos to wireless VR headsets. We spatially partition the underlying 3D mesh into multiple 3D sub-meshes, and construct an efficient 3D geometry mesh called hexaface sphere to optimally represent tiled 360 VR videos in the 3D space. We then spatially divide the 360 videos into multiple tiles while encoding and packaging, use MPEG-DASH SRD to describe the spatial relationship of tiles in the 3D space, and prioritize the tiles in the Field of View (FoV) for view-aware adaptation. Our initial evaluation results show that we can save up to 72% of the required bandwidth on 360 VR video streaming with minor negative quality impacts compared to the baseline scenario when no adaptations is applied.

Multimedia

What problem does this paper attempt to address?

This paper aims to solve the problem of high - bandwidth requirements in 360 - degree virtual reality (VR) video streaming, especially for applications on wireless VR head - mounted displays (HMDs). As 4K video resolution has become the basic requirement for current HMDs, and the demand for even higher resolutions such as 8K is increasing day by day, how to efficiently transmit these large - scale 360 - degree VR videos in the bandwidth - limited wireless network environment and maintain an acceptable quality level has become a major challenge.

The author proposes an adaptive bandwidth - efficient 360 - degree VR video streaming system based on MPEG - DASH SRD. By extending MPEG - DASH SRD to 3D space, the author develops a dynamic view - aware adaptive technique to reduce the high bandwidth required for 360 - degree VR video streaming to wireless VR headsets. Specific methods include:

1. **Spatial Partitioning**: Divide the underlying 3D grid into multiple 3D sub - grids and construct an efficient 3D geometric grid called "hexaface sphere" to optimally represent the tiling in 360 - degree VR videos.
2. **Encoding and Packaging**: Divide the 360 - degree video space into multiple tiles during the encoding and packaging process.
3. **Describing Spatial Relationships**: Use MPEG - DASH SRD to describe the spatial relationships of tiles in 3D space.
4. **Priority Allocation**: Give priority to tiles within the Field of View (FoV) for view - aware adaptive processing, thereby optimizing bandwidth usage.

Preliminary evaluation results show that compared with the baseline situation without applying any adaptive techniques, this method can save up to 72% of the required bandwidth while having a relatively small impact on video quality. This indicates that even in the case of limited bandwidth, users can still obtain a high - quality immersive experience.