List All Projects List Graduate Projects List Senior Design Projects List REU Projects Other Undergraduate Projects
Project # 81
Tile-Based Adaptive Bitrate Streaming for Bandwidth-Efficient VR Video Delivery
| Team Lead: | John Cobio |
| Other Team Members: | |
| Faculty Advisor(s): | Ming Li |
| Department: | CSE |
| Project Type: | REU |
| Title: | Tile-Based Adaptive Bitrate Streaming for Bandwidth-Efficient VR Video Delivery |
| Abstract: | Virtual reality video streaming demands extremely high bandwidth to deliver immersive, high-quality experiences, often exceeding the capacity of current network infrastructure. This research presents a tile-based adaptive bitrate streaming approach that significantly reduces bandwidth consumption while maintaining high visual quality for VR video content. By spatially dividing each video frame into independent tiles, the system selectively allocates higher bitrates to tiles within the user's predicted viewport and lower bitrates to peripheral regions, exploiting the natural limitations of human visual attention in head-mounted displays. Our implementation, built on a Unity-based VR platform, dynamically adjusts tile quality using real-time viewport prediction and network condition monitoring. Experimental evaluation demonstrates substantial bandwidth savings compared to traditional full-frame streaming methods, where every portion of the 360-degree video is delivered at uniform quality regardless of where the user is looking. The tile-based approach enables high-resolution VR video delivery over bandwidth-constrained networks without perceptible quality degradation in the user's active field of view. This work contributes to making VR video streaming more practical and scalable for consumer applications by addressing one of the key bottlenecks in immersive media delivery: the inefficient use of network resources. |
| Poster: | View Poster #81 (PDF) |
| Judges Scoring: | Submit Evaluation (judges only) |