Omnidirectional Media Track Mapping for Low-Overhead Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions for encapsulating and transmitting omnidirectional media content are complex and not fully compliant with existing mechanisms, particularly in signaling tracks and mapping between original omnidirectional media content and two-dimensional media contents, leading to overhead and inefficiencies.
Innovation Solution
A method for encapsulating and streaming omnidirectional media content using the ISO Base Media File Format, incorporating SpatialRelationship2DdescriptionBox and source_id to manage spatial relationships between tracks, allowing for efficient parsing and rendering on client devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If existing solutions for encapsulating omnidirectional media content are used, then the content can be transmitted, but the signaling and encapsulation process becomes complex with overhead
Solution Approach 1:
The patent segments omnidirectional media content into multiple sub-picture tracks, each representing a specific spatial region. This segmentation allows independent encoding and transmission of different spatial portions, simplifying the overall encapsulation process while reducing signaling overhead compared to transmitting complete omnidirectional content in a single track.
Solution Approach 2:
The patent applies local quality by assigning specific spatial relationship metadata (SpatialRelationship2DdescriptionBox) to individual sub-picture tracks based on their spatial characteristics. Each track receives tailored metadata describing its position, size, and orientation relative to the original omnidirectional content, rather than applying uniform complex signaling to all tracks.
2Productivity
If existing solutions for mapping between omnidirectional and two-dimensional media contents are used, then the mapping can be achieved, but overhead and inefficiencies occur
Solution Approach 1:
The patent performs preliminary action by pre-calculating and embedding spatial relationship metadata (position, size, orientation) directly into each sub-picture track during the encapsulation phase. This preliminary organization of spatial information eliminates the need for complex runtime calculations during parsing and rendering, thereby improving efficiency and reducing transmission overhead.
Solution Approach 2:
The patent introduces SpatialRelationship2DdescriptionBox as an intermediary element that mediates between the original omnidirectional media content and the two-dimensional sub-picture tracks. This intermediary structure efficiently carries essential spatial mapping information without requiring transmission of redundant data, reducing overhead while enabling accurate reconstruction.
3Reliability
If multiple sub-picture tracks are used to represent omnidirectional content, then spatial relationships can be managed, but signaling overhead increases
Solution Approach 1:
The patent applies universality by designing the SpatialRelationship2DdescriptionBox to serve multiple functions simultaneously: it encodes position, size, orientation, and spatial mapping relationships in a single compact metadata structure. This multi-functional approach maintains accurate spatial relationships across multiple sub-picture tracks while minimizing signaling overhead through consolidated information representation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for encapsulating encoded media data corresponding to a wide view of a scene, the method comprising: obtaining a projected picture from the wide view of the scene; splitting the projected picture into at least one sub-picture; encoding the at least one sub-picture into a plurality of tracks; generating descriptive metadata associated with the encoded tracks, the descriptive metadata comprise a first item of information associated with each track being indicative of a spatial relationship between the at least one sub-picture encoded in the track and a reference picture; wherein the descriptive metadata further comprises a second item of information indicating the reference picture.