Non-Dyadic Block Transform Scaling and Quantization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards face challenges in efficiently processing blocks with non-dyadic dimensions, as existing transform and quantization mechanisms are optimized for dyadic blocks, leading to suboptimal performance when applied to non-dyadic blocks.
Innovation Solution
The proposed solution involves modifying the scaling and quantization processes for non-dyadic blocks by applying specific scaling factors and shifting values during forward and inverse transforms, and adjusting quantization parameters based on block dimensions, allowing for effective compression and decompression of non-dyadic blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If existing transform and quantization mechanisms optimized for dyadic blocks are applied to non-dyadic blocks, then device complexity is reduced by using standardized processing, but coding performance deteriorates due to suboptimal processing of non-dyadic blocks
Solution Approach 1:
The patent applies different processing approaches based on block type: dyadic blocks use standard transform and quantization mechanisms, while non-dyadic blocks use modified mechanisms with adjusted scaling factors and quantization parameters. This local differentiation optimizes coding performance for each block type without unnecessarily increasing overall device complexity.
Solution Approach 2:
The patent modifies scaling factors and quantization parameters specifically for non-dyadic blocks. By changing these parameters rather than the fundamental processing architecture, the patent improves coding performance while keeping device complexity increases minimal.
2Ease of operation
If standardized transform and quantization processes are used for all blocks, then ease of operation is improved by simplifying processing, but productivity deteriorates due to suboptimal compression efficiency for non-dyadic blocks
Solution Approach 1:
The patent segments the processing into two paths: one for dyadic blocks using standardized processes, and another for non-dyadic blocks using modified processes. This segmentation allows the system to maintain simplicity for the majority case while optimizing for the specific case without significantly complicating overall operation.
Solution Approach 2:
The patent applies modified scaling and quantization only partially - specifically for non-dyadic blocks where it is needed - rather than applying complex processing universally. This partial action maintains ease of operation for standard cases while improving productivity where required.
3Manufacturing precision
If modified scaling and quantization processes are applied to non-dyadic blocks, then coding performance is improved through tailored processing, but device complexity increases due to additional processing logic
Solution Approach 1:
The patent improves coding performance for non-dyadic blocks by modifying parameters (scaling factors, quantization values) rather than changing the fundamental processing architecture. This approach achieves better performance while minimizing increases in device complexity.
Solution Approach 2:
The patent designs the modified processing to handle both dyadic and non-dyadic blocks within a unified framework. The same transform and quantization infrastructure serves both block types, with parameter adjustments applied conditionally, thereby reducing the complexity increase from handling multiple block types.
Data Source
AI summary
A mechanism for processing video data is disclosed. A scaling process is selected for application to a block during residual coding based on whether the block is dyadic or non-dyadic. The block has a width (W) and a height (H). A conversion is performed between a visual media data and a bitstream based on application of the scaling process to the block.