HEVC Reference Layer Configuration for Scalable Video
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital video coding technologies face challenges in efficiently encoding and decoding video signals across heterogeneous environments with varying network conditions and device capabilities, leading to suboptimal video quality and increased computational complexity.
Innovation Solution
A video coding device that employs a multi-loop SHVC encoder with a processor to limit the number of reference layers, using a scalable video coding system to encode video signals into layers with constrained inter-layer dependencies, and a video decoding device that determines video representation formats and includes direct dependency type information to optimize bitstream extraction and decoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of reference layers is increased to improve video quality and enable more flexible decoding, then video quality and adaptability are improved, but coding complexity and computational resources increase
Solution Approach 1:
The patent applies parameter changes by introducing a maximum number of reference layers parameter (NumMaxRefLayers) that can be configured at different levels (VPS, SPS, or PPS). This parameter controls the number of reference layers that can be used for decoding, allowing the system to adapt the complexity level according to available computational resources while maintaining the ability to use more reference layers when higher quality is needed. The parameter can be set to -1 to indicate no limit, or to a specific value to constrain the number of reference layers.
Solution Approach 2:
The patent implements dynamics by making the reference layer configuration adaptive rather than fixed. The NumMaxRefLayers parameter can be changed at different initialization points (VPS, SPS, or PPS) and can be adjusted based on network conditions, device capabilities, and quality requirements. This allows the system to dynamically adapt the coding complexity and video quality trade-off in real-time.
2Adaptability or versatility
If more reference layers are used to enable efficient decoding across heterogeneous environments, then adaptability is improved, but computational resources and processing time increase
Solution Approach 1:
The patent enables adaptability through the NumMaxRefLayers parameter that can be configured according to device capabilities and network conditions. The parameter can be set at VPS level for all layers, at SPS level for specific sequences, or at PPS level for specific pictures, allowing fine-grained control over computational resource usage while maintaining adaptability to heterogeneous decoding environments.
Solution Approach 2:
The patent segments the control of reference layer configuration into different levels (VPS, SPS, PPS), allowing different parts of the bitstream to have different maximum reference layer settings. This segmentation enables the system to apply different computational resource constraints to different video sequences or pictures based on their specific requirements and the available decoding resources.
3Device complexity
If the maximum number of reference layers is constrained to reduce coding complexity, then device complexity is reduced, but video quality and decoding flexibility may be compromised
Solution Approach 1:
The patent resolves this contradiction by making the NumMaxRefLayers parameter configurable rather than fixed. The parameter can be set to -1 (no limit) when high video quality is required and computational resources are abundant, or set to a specific value when coding complexity needs to be reduced. The ability to change this parameter at different initialization points allows the system to optimize the trade-off between video quality and device complexity based on specific application requirements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A video coding device may identify a network abstraction layer (NAL) unit. The video coding device may determine whether the NAL unit includes an active parameter set for a current layer. When the NAL unit includes the active parameter set for the current layer, the video coding device may set an NAL unit header layer identifier associated with the NAL unit to at least one of: zero, a value indicative of the current layer, or a value indicative of a reference layer of the current layer. The NAL unit may be a picture parameter set (PPS) NAL unit. The NAL unit may be a sequence parameter set (SPS) NAL unit.