DASH MPD Viewpoint Descriptors for VR Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Dynamic Adaptive Streaming over HTTP (DASH) mechanisms face challenges in efficiently streaming Virtual Reality (VR) video due to its higher data requirements and complex viewpoint management, including large MPD files from multiple viewpoints, lack of clear static and dynamic viewpoint descriptors, and inadequate association of timed metadata with entire media presentations.
Innovation Solution
The introduction of viewpoint descriptors that indicate static or dynamic viewpoints, association IDs for efficient object referencing, and the omission of unnecessary attributes in timed metadata representations to reduce file size and improve association efficiency, allowing for more streamlined VR video streaming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional DASH mechanisms are used to stream VR video, then video streaming functionality is provided, but MPD file size becomes excessively large due to multiple viewpoints and inadequate viewpoint management
Solution Approach 1:
The patent segments viewpoint information into distinct descriptors (VWPT descriptors) that can be independently managed and associated with specific media content. Each viewpoint is represented as a separate entity with its own identifier and position information, allowing the system to handle multiple viewpoints without proportionally increasing overall descriptor complexity.
Solution Approach 2:
The patent introduces a new dimension for organizing VR content by adding viewpoint-specific descriptors that operate alongside traditional DASH structure. The VWPT descriptors add a viewpoint dimension to the existing media presentation hierarchy, enabling efficient organization of spherical video content without flattening the structure into a single large MPD file.
2Adaptability or versatility
If comprehensive viewpoint information is included in MPD descriptors, then viewpoint management capability is improved, but file size and processing complexity increase
Solution Approach 1:
The patent extracts viewpoint-specific information into separate VWPT descriptors that are associated with media content through references rather than being embedded directly in every MPD element. This separation allows comprehensive viewpoint data to be maintained while reducing the complexity burden on the core DASH structure.
Solution Approach 2:
The VWPT descriptor structure is designed to be universal and applicable to different types of VR content and viewpoints. The same descriptor framework handles both static and dynamic viewpoints, as well as different coordinate systems, without requiring separate specialized structures for each case.
3Measurement precision
If timed metadata is associated with every media track individually, then metadata precision is improved, but association efficiency and file size deteriorate
Solution Approach 1:
The patent merges timed metadata associations at the media presentation level rather than requiring individual track-level associations. The VWPT descriptors can reference multiple media tracks simultaneously, allowing common viewpoint information to be shared across multiple tracks without redundant declarations, thus improving association efficiency while maintaining precision.
Data Source
AI summary
A method comprises obtaining media content associated with a first viewpoint and a second viewpoint, wherein a first position of the first viewpoint is static and a second position of the second viewpoint is dynamic; generating a first VWPT descriptor based on the media content and describing the first viewpoint; generating a second VWPT descriptor based on the media content and describing the second viewpoint; generating an MPD comprising the first VWPT descriptor and the second VWPT descriptor; and transmitting the MPD.