Adaptive Sub-block Quantization in Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video encoding methods lack fine control over quantization, as it is performed with a single parameter for each macroblock, failing to adapt to varying image properties within a macroblock, even though motion compensation and orthogonal transforms can be done in smaller, variable-sized blocks.
Innovation Solution
Implementing a quantization parameter switching mechanism that allows for finer control by changing quantization parameter values in units of sub-blocks, enabling the use of different quantization parameters for sub-blocks within a macroblock based on their activity variance, thereby allowing for adaptive quantization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If quantization is performed with a single parameter for each macroblock, then device complexity is reduced, but manufacturing precision of quantization control deteriorates
Solution Approach 1:
The macroblock is divided into multiple sub-blocks, and each sub-block is assigned its own quantization parameter. This segmentation allows independent control of quantization for different regions within a macroblock, enabling fine-grained adaptation to local image characteristics while maintaining a relatively simple overall control structure.
Solution Approach 2:
Different quantization parameters are applied to different sub-blocks based on their local image properties such as activity variance. This allows the quantization process to adapt locally to varying image characteristics within each macroblock, improving precision without requiring a completely complex global control system.
2Manufacturing precision
If quantization parameter values are changed in units of sub-blocks, then manufacturing precision of quantization control is improved, but device complexity increases
Solution Approach 1:
The macroblock is divided into multiple sub-blocks, and each sub-block is assigned its own quantization parameter. This segmentation allows independent control of quantization for different regions within a macroblock, enabling fine-grained adaptation to local image characteristics while maintaining a relatively simple overall control structure.
Solution Approach 2:
The quantization parameter is made dynamic by allowing it to vary across different sub-blocks within a macroblock based on local image properties. This dynamic approach enables adaptive quantization control that responds to local variations in image content, improving precision without requiring a completely complex static control system.
3Ease of operation
If a single quantization parameter is used for each macroblock, then ease of operation is improved, but adaptability to varying image properties within macroblock deteriorates
Solution Approach 1:
The macroblock is divided into multiple sub-blocks, and each sub-block is assigned its own quantization parameter. This segmentation allows independent control of quantization for different regions within a macroblock, enabling fine-grained adaptation to local image characteristics while maintaining a relatively simple overall control structure.
Solution Approach 2:
Different quantization parameters are applied to different sub-blocks based on their local image properties such as activity variance. This allows the quantization process to adapt locally to varying image characteristics within each macroblock, improving precision without requiring a completely complex global control system.
4Adaptability or versatility
If quantization parameter values are changed in units of sub-blocks, then adaptability to varying image properties is improved, but ease of operation deteriorates
Solution Approach 1:
The macroblock is divided into multiple sub-blocks, and each sub-block is assigned its own quantization parameter. This segmentation allows independent control of quantization for different regions within a macroblock, enabling fine-grained adaptation to local image characteristics while maintaining a relatively simple overall control structure.
Solution Approach 2:
The quantization parameter is made dynamic by allowing it to vary across different sub-blocks within a macroblock based on local image properties. This dynamic approach enables adaptive quantization control that responds to local variations in image content, improving precision without requiring a completely complex static control system.
Data Source
AI summary
To allow a finer quantization control according to the property of an image within a macroblock, quantization parameter values are allowed to be changed in units of sub-blocks equal to or smaller than the macroblock in a similar manner as in motion compensation and orthogonal transform processes. A finer-tuned quantization control is performed, for example, by selecting fine and coarse quantization parameters respectively for corresponding sub-blocks if a plurality of images having different properties coexist within the macroblock.