Quality Track Segmentation for Adaptive Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing media content delivery systems face challenges in dynamically adapting to varying network conditions and client resources, leading to fluctuations in media content quality due to the lack of effective quality information management.
Innovation Solution
A mechanism for carrying quality information of media content is introduced, where quality metrics are computed and stored in a separate quality track within the ISO-BMFF format, enabling a one-to-one mapping with encoded media data, allowing clients to make informed streaming adaptation decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If quality information is embedded within media data tracks, then quality metrics are readily available for adaptive streaming, but the media file structure becomes more complex and processing overhead increases
Solution Approach 1:
The patent segments quality information from media data by creating separate quality tracks that are independently structured. Each quality track contains only quality metrics (PSNR, SSIM, etc.) corresponding to specific media segments, allowing quality information to be processed separately from the complex media data while maintaining precise associations through segment identifiers.
Solution Approach 2:
The patent extracts quality information from the media data stream and places it in separate quality tracks. This extraction eliminates the need to parse through complex media data structures to access quality metrics, as quality information is independently organized and可直接 accessed through the quality track structure.
2Measurement precision
If multiple quality metrics are computed and stored for different granularity levels, then adaptive streaming decisions are more accurate, but the amount of quality information to be processed and transmitted increases
Solution Approach 1:
The patent implements a dynamic quality information structure where multiple granularity levels (segment, sub-segment, sample levels) are organized hierarchically. Clients can dynamically select which granularity level to use based on their specific needs and network conditions, allowing accurate quality assessment without necessarily processing all quality metrics at all levels.
Solution Approach 2:
The patent merges quality information from multiple granularity levels into a unified quality track structure. Instead of creating separate files or data structures for each granularity level, all quality metrics are integrated into a single organized structure that can be efficiently accessed and filtered based on client requirements.
3Ease of operation
If quality tracks are stored separately from media data tracks, then quality information can be independently managed and processed, but the association between quality metrics and corresponding media data must be maintained
Solution Approach 1:
The patent establishes a feedback mechanism where quality tracks continuously reference back to their corresponding media data tracks through segment identifiers and time synchronization. This ensures that quality metrics remain correctly associated with the media segments they describe, enabling independent management while maintaining accurate associations through systematic cross-referencing.
Data Source
AI summary
A method for managing quality information for dynamic adaption in content delivery, wherein the method comprises selecting a granularity level for an encoded media data, computing a plurality of quality metrics for the encoded media data at the selected granularity level, wherein the quality metrics are measures of coding quality of the encoded media data, generating a quality track comprising the plurality of quality metrics, and storing the quality track on a media content server separate from other tracks that are associated with the encoded media data. Also disclosed is a network device comprising a processor configured to select a granularity level for an encoded media data, and compute a plurality of quality metrics that indicate coding quality of the encoded media data at the selected granularity level, and a transmitter coupled to the processor and configured to send the quality track towards a user of the media data.