Multi-layer Video Coding Access Unit Delimiter Semantics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-layer video coding technologies face challenges in determining the representation format for independent non-base layers (INBLs) and interpreting Supplemental Enhancement Information (SEI) messages across different layers, as well as processing Access Unit Delimiters (AUD) NAL units, which are not clearly defined in existing standards.
Innovation Solution
The proposed solution involves generating specific Network Abstraction Layer (NAL) units for different pictures within an access unit, determining the appropriate Video Parameter Sets (SPS) or Picture Parameter Sets (PPS) for interpreting SEI messages, and modifying the semantics of AUD NAL units to clearly indicate the type of slices present in a coded picture, ensuring compatibility across layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing standards are used for multi-layer video coding, then basic video compression is achieved, but the representation format for independent non-base layers and interpretation of SEI messages across different layers are not clearly defined
Solution Approach 1:
The patent segments the video coding structure into distinct layers (base layer and independent non-base layers) with clearly defined NAL unit types for each layer. This segmentation allows each layer to be processed independently while maintaining overall standard compliance, resolving the ambiguity in representing multi-layer video data.
Solution Approach 2:
The patent introduces an intermediary mechanism through modified AUD NAL units that carry layer-specific information. These AUD units act as mediators between different layers, providing the necessary information for interpreting SEI messages and determining representation formats without requiring complex cross-layer dependencies.
2Adaptability or versatility
If access unit delimiter semantics are modified to indicate slice types in multi-layer coding, then compatibility across layers is improved, but processing complexity increases
Solution Approach 1:
The patent extends the universality of AUD NAL units by enabling them to serve multiple functions: maintaining their original access unit boundary indication role while simultaneously carrying slice type information for multi-layer compatibility. This multi-functionality improves layer compatibility without requiring separate structures.
Solution Approach 2:
The patent performs preliminary action by embedding slice type information in AUD NAL units at the access unit level, before detailed slice processing occurs. This allows decoders to pre-determine compatibility requirements and prepare appropriate processing paths, reducing actual processing complexity during decoding.
3Productivity
If independent non-base layers are encoded with clear representation formats, then decoding efficiency is improved, but encoding complexity increases
Solution Approach 1:
The patent applies parameter changes by introducing specific NAL unit type parameters for independent non-base layers and modifying AUD NAL unit parameters to include slice type information. These parameter changes enable clear representation formats that improve decoding efficiency while keeping encoding complexity manageable through standardized parameter structures.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A device for processing video data includes a memory configured to store at least a portion of a bitstream of multi-layer video data and one or more processors configured to generate a first video coding layer (VCL) network abstraction layer (NAL) unit for a first picture of an access unit, the first VCL NAL unit comprising a first slice type; generate a second VCL NAL unit for a second picture of the access unit, the second VCL NAL unit comprising a second slice type; and generate an access unit delimiter (AUD) NAL unit based on the first and second slice types.