Media File Encapsulation Using Sample Duplication for Time Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The encapsulation of media data into fragmented media files is challenged by time boundaries that do not align with the presentation duration of samples, particularly when dealing with sparse samples or samples with dynamic content, leading to issues in starting at specific times or seeking within the media presentation.
Innovation Solution
The method involves duplicating original samples into two consecutive samples separated by a time boundary, with a time adjustment indication in the metadata to maintain the intended rendering, ensuring alignment and synchronization across fragmented tracks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If media data is encapsulated into fragmented media files with time boundaries, then the media file can support random access and seeking operations, but time boundaries may cross sample presentation durations causing rendering timing issues
Solution Approach 1:
The original sample is segmented into multiple duplicated samples at the time boundary points. Each duplicated sample represents a portion of the original sample's presentation duration, allowing the fragmented media file to maintain proper timing alignment while supporting random access operations.
Solution Approach 2:
The original sample is copied into multiple duplicated samples with adjusted timing information. The copy includes metadata that indicates the relationship between the duplicated samples and the original sample, preserving the intended rendering timing while adapting to the fragmented structure.
2Manufacturing precision
If original samples are duplicated at time boundaries, then timing alignment across fragmented tracks is maintained, but the number of samples increases leading to larger file size
Solution Approach 1:
Multiple duplicated samples that represent portions of the same original sample are merged during playback by the rendering device. The metadata indicates which samples are duplicates of the same original sample, allowing the system to combine them back into the intended continuous presentation, thus reducing the effective sample count while maintaining timing precision.
3Manufacturing precision
If time adjustment indication is added to metadata, then rendering timing is preserved across fragments, but the metadata complexity increases
Solution Approach 1:
The time adjustment information is pre-calculated and embedded in the metadata during the encapsulation process. This preliminary action stores the relationship between original and duplicated samples in advance, allowing the rendering device to simply read and apply the timing adjustments without performing complex calculations, thus reducing runtime complexity.
Data Source
AI summary
A method of encapsulating media data into a media file, the method comprising: obtaining a time boundary in a time interval corresponding to the presentation duration of an original sample of the media data; duplicating the original sample into two consecutive duplicated samples separated by the time boundary, the two duplicated samples having a same data part corresponding to the data part of the original sample; generating a time adjustment indication indicating for each duplicated sample the relationship between the time interval corresponding to the presentation duration of the duplicated sample and the time interval corresponding to the presentation duration of the original sample; and encapsulating the media data comprising the duplicated samples and the time adjustment indication in the media file, the time adjustment indication being located in metadata describing the media data.