NAL Unit Types for Broken Link Picture Indication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques face inefficiencies in random access and stream adaptation due to the need for signaling 'broken link' flags in the slice header or adaptation parameter sets, which require complex entropy encoding and decoding, and do not effectively handle non-decodable leading pictures in clean random access (CRA) pictures.
Innovation Solution
The use of distinct Network Abstraction Layer (NAL) unit types to indicate the presence of broken link pictures, allowing for decodable or non-decodable leading pictures without the need for entropy encoding or decoding, simplifying the process of identifying and converting normal CRA pictures to broken link CRA (BLC) pictures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If broken link flags are signaled in the slice header or adaptation parameter sets, then the presence of broken link pictures can be indicated, but the complexity of entropy encoding and decoding increases
Solution Approach 1:
The patent extracts the broken link indication from the complex entropy-coded slice header or adaptation parameter sets and places it in the simpler NAL unit header. This separates the critical broken link flag from the entropy-coded data, allowing it to be read without performing complex entropy decoding operations.
Solution Approach 2:
The NAL unit header serves as an intermediary layer between the video data and the broken link indication. By placing the broken link flag in this intermediate structure that is already parsed during decoding, the patent avoids the need for additional entropy decoding steps while ensuring the information is available to the decoder.
2Adaptability or versatility
If normal CRA pictures are converted to broken link CRA (BLC) pictures, then stream adaptation and splicing are enhanced, but the process requires complex entropy encoding and decoding
Solution Approach 1:
The patent performs the broken link indication setup in advance by placing the flag in the NAL unit header before the entropy-coded slice data. This preliminary placement of the flag allows receiving devices to identify BLC pictures without having to perform complex entropy decoding to determine the picture type, enabling faster stream adaptation and splicing operations.
3Adaptability or versatility
If leading pictures are made non-decodable in CRA pictures, then stream splicing becomes more flexible, but the handling complexity increases
Solution Approach 1:
The patent extracts the critical identification function from the complex slice header structure and places it in the NAL unit header. This allows receiving devices to quickly identify pictures with non-decodable leading pictures by checking the NAL unit type and broken link flag without having to parse and decode the entire slice header structure, reducing handling complexity while maintaining splicing flexibility.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
Various techniques for enhanced support of stream adaptation and splicing based on clean random access (CRA) pictures are described. Instead of using a flag in the slice header to indicate that a broken link picture is present, a distinct network abstraction layer (NAL) unit type can be used to indicate the presence of a broken link picture. In some implementations, a first distinct NAL unit type may be used to indicate the presence of a broken link picture with leading pictures, while a second distinct NAL unit type indicates the presence of a broken link picture without leading pictures. In some implementations, a third distinct NAL unit type may be used to indicate the presence of a broken link picture with decodable leading pictures.