Predicting Field of View for High-Latency Panoramic Video Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In live panoramic video presentations, users with higher latency experience delayed content delivery, leading to inefficiencies as they cannot view all content due to their limited field of view, while other users with lower latency can view the entire scene, resulting in unequal viewing experiences.
Innovation Solution
A system predicts the field of view of high-latency users by analyzing data from lower-latency users, generating a predicted viewing map, and adjusts the video presentation to omit or reduce the quality of content outside the predicted field of view for high-latency users, ensuring a better viewing experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video content is delivered to all users with full quality and complete scene coverage, then lower-latency users can view the entire scene, but high-latency users experience delayed content delivery and cannot view all content effectively due to their limited field of view
Solution Approach 1:
The system applies different quality levels and content regions to different user groups based on their latency characteristics. High-latency users receive video content optimized for their specific predicted field of view with higher quality, while other regions are delivered at reduced quality or omitted entirely. This local quality differentiation resolves the contradiction by ensuring high-latency users receive sufficient content quality within their effective viewing window without requiring complete scene coverage
Solution Approach 2:
The system performs preliminary field of view prediction for high-latency users before delivering video content. By analyzing the movement patterns and viewing behavior of lower-latency users, the system predicts where high-latency users will be looking and pre-prepares optimized video segments for those specific regions. This preliminary action allows the system to proactively address the latency issue by delivering relevant content in advance rather than waiting for user requests
2Adaptability or versatility
If complete panoramic video content is delivered to all users, then all users can potentially view the entire scene, but bandwidth usage increases and high-latency users still cannot effectively utilize the content outside their field of view
Solution Approach 1:
The system delivers only the necessary portion of the panoramic video content to high-latency users based on their predicted field of view, rather than delivering the complete 360-degree scene. This partial action approach provides sufficient content for high-latency users to have an effective viewing experience within their latency constraints, while avoiding the excessive bandwidth consumption that would result from delivering complete panoramic content to all users
Solution Approach 2:
The system dynamically changes video delivery parameters including resolution, bitrate, and content region based on user latency characteristics and predicted field of view. For high-latency users, the system adjusts these parameters to optimize delivery efficiency by focusing resources on the predicted viewing regions rather than uniformly delivering complete scene coverage. This parameter adaptation resolves the contradiction by making bandwidth usage proportional to actual user needs
Data Source
AI summary
Aspects of the subject disclosure may include, for example, a method including providing a video presentation to user devices. A latency related to the video presentation is determined for each device; a high-latency user device is identified having greater latency than other, low-latency devices. Field of view information is analyzed for each of the low-latency devices; based on the analyzing, a field of view of the high-latency user device is predicted. In accordance with the predicting, a predicted viewing map for the high-latency user device is generated. The video presentation for the high-latency user device is adjusted based on the predicted viewing map. Other embodiments are disclosed.