MMT Stream Timing Correction for Parallel UHD Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Decoding ultra-high definition moving images such as 8K and 4K in real time is challenging due to high processing loads, and existing methods struggle with demultiplexing and decoding efficiency, especially when leap second adjustments occur, leading to delayed or incorrect presentation of data units.
Innovation Solution
The method involves dividing access units into slice segments and associating them with tiles, using a multiplexing format like MMT, and employing multiple decoders synchronized by a reference clock to ensure real-time decoding and presentation, even with leap second adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple decoders are used to decode ultra-high definition moving images in real time, then the processing load is reduced and real-time decoding is achieved, but the device complexity increases
Solution Approach 1:
The patent divides the access unit into multiple slice segments and associates each slice segment with a separate decoder. This segmentation allows the decoding process to be distributed across multiple decoders, reducing the processing load on each individual decoder and enabling real-time decoding of ultra-high definition moving images while managing device complexity through structured division of labor
2Reliability
If leap second adjustment is made to reference time information, then time synchronization is maintained, but the presentation timing of data units becomes delayed or incorrect
Solution Approach 1:
The patent performs preliminary action by detecting leap second adjustment information in advance and proactively correcting the presentation timing of data units before the timing error occurs. The receiving device monitors reference time information, detects leap second adjustments, and applies timing corrections to the associated data units ahead of time, ensuring that presentation timing remains accurate despite leap second events
Solution Approach 2:
The patent implements a feedback mechanism where the receiving device continuously monitors reference time information for leap second adjustment indicators, detects when such adjustments occur, and uses this feedback to dynamically correct the presentation timing of data units. This closed-loop approach ensures time synchronization is maintained while preventing timing delays or errors in data unit presentation
3Productivity
If access units are divided into slice segments and associated with tiles, then decoding efficiency is improved, but the demultiplexing complexity increases
Solution Approach 1:
The patent segments access units into multiple slice segments and associates each slice segment with specific tiles, enabling parallel decoding operations. This segmentation improves decoding efficiency by allowing multiple decoders to work simultaneously on different slice segments while maintaining the spatial structure through tile associations
Solution Approach 2:
The patent applies local quality by associating specific slice segments with specific tiles based on their spatial characteristics. Each slice segment is processed with considerations for its local spatial structure and tile associations, allowing optimized decoding for different regions of the image while improving overall decoding efficiency
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A receiving method of receiving a first data unit in which data making up an encoded stream is stored and the first data unit stores a plurality of second data units. The receiving method includes: receiving the first data unit, first time information indicating a presentation time of the first data unit, second time information indicating, together with the first time information, a presentation time or a decoding time of each of the plurality of second data units, and identification information; calculating the presentation time or the decoding time of each of the plurality of second data units using the first time information and the second time information; and correcting the presentation time or the decoding time of each of the plurality of second data units based on the identification information.