Decoded Picture Buffer Allocation for Scalable Video Layer Output
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face challenges in efficiently managing decoded picture buffer (DPB) memory allocation and output of decoded pictures, particularly in scalable video coding, which can lead to inefficiencies and complexity when handling multiple video layers.
Innovation Solution
The proposed methods involve converting video layers into coded representations that adhere to specific format rules, including controlling DPB output based on maximum picture dimensions, bit depth, and chroma format indicators, and using format rules that are independent of separate color planes or indicated in access units, to optimize DPB management and picture output.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple video layers are processed in scalable video coding, then video quality and adaptability are improved, but DPB memory allocation complexity and device complexity increase
Solution Approach 1:
The patent divides the decoded picture buffer into separate buffer structures for different video layers, allowing independent management of each layer's pictures. This segmentation enables the decoder to handle multiple layers efficiently by treating them as distinct entities with separate output timing controls, thereby reducing the overall complexity of DPB management in scalable video coding.
Solution Approach 2:
The patent introduces dynamic control mechanisms where the output timing of pictures from the DPB is adjusted based on layer-specific parameters and decoding requirements. Each video layer can have its own output timing configuration, allowing the system to adaptively manage buffer operations according to the specific needs of each layer, thus improving scalability while maintaining manageable complexity.
2Adaptability or versatility
If separate color planes are used for encoding, then chroma format flexibility is improved, but the variable indicating DPB output becomes dependent on color plane configuration
Solution Approach 1:
The patent extracts the DPB output control variable from its dependency on separate color plane configuration. By decoupling the output timing control from the chroma format indication, the system maintains flexibility in chroma format selection while simplifying the DPB management logic. The output control is now determined by layer and picture parameters rather than color plane configuration, reducing complexity.
3Manufacturing precision
If maximum picture width and height for all video layers control DPB output, then memory allocation accuracy is improved, but the control mechanism becomes more complex
Solution Approach 1:
The patent establishes maximum picture width and height parameters for all video layers in advance, before actual decoding and buffer operations. These pre-defined maximum dimensions are used to configure the DPB capacity and allocation strategy upfront, ensuring accurate memory allocation while simplifying runtime control. The preliminary configuration eliminates the need for complex dynamic adjustments during decoding.
4Manufacturing precision
If chroma format indicator and bit depth maximum values are controlled, then picture representation precision is improved, but DPB management complexity increases
Solution Approach 1:
The patent applies different chroma format and bit depth configurations to different video layers based on their specific requirements. Each layer can have its own maximum chroma format indicator and bit depth values, allowing high precision where needed while maintaining efficiency elsewhere. This localized quality control prevents unnecessary complexity by only applying stringent parameters where they are actually required for picture representation.
Data Source
AI summary
Methods and apparatus for video processing, including coding and decoding, are described. One example video processing method includes performing a conversion between a video having one or more video layers and a bitstream of the video according to a format rule, and wherein the format rule specifies that a value of a flag that indicates whether to remove pictures previously decoded and stored in a decoded picture buffer from the decoded picture buffer when decoding an access unit of a certain type is included in the bitstream.