2D Separable Transform Transpose Buffer Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing size of transforms in video coding standards like HEVC leads to higher complexity in hardware and software implementations due to the large transpose buffer requirements, which is costly and inefficient.
Innovation Solution
Implementing a low-complexity 2D separable transform design with variable bit width intermediate results, allowing for reduced bit width of intermediate results stored in a transpose buffer, thereby reducing the required buffer size and simplifying arithmetic units, while maintaining performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large transform sizes (e.g., 16×16 and 32×32) are used in HEVC, then video coding performance is improved, but hardware and software implementation complexity increases due to large transpose buffer requirements
Solution Approach 1:
The patent applies parameter changes by using variable bit width for intermediate results instead of fixed bit width. Different intermediate results are allocated different bit widths based on their actual precision requirements, allowing the system to maintain high video coding performance while reducing the overall buffer size and implementation complexity
Solution Approach 2:
The patent implements local quality by assigning different precision levels to different intermediate results based on their specific needs. Critical intermediate results that require high precision for video quality are allocated more bits, while less critical results use fewer bits, optimizing the trade-off between performance and complexity
2Ease of manufacture
If fixed bit width is used for all intermediate results, then implementation is simpler, but buffer size increases and resources are wasted
Solution Approach 1:
The patent introduces dynamics by making the bit width of intermediate results variable rather than fixed. The system dynamically adjusts the bit width allocation based on the specific requirements of each intermediate result, enabling more efficient resource utilization while maintaining implementation feasibility
Solution Approach 2:
The patent changes the parameter of bit width from a fixed value to a variable parameter that adapts to the needs of different intermediate results. This parameter change allows the system to reduce total buffer size while maintaining the simplicity of fixed-width operations for individual results
Data Source
AI summary
Methods are provided for reducing the size of a transpose buffer used for computation of a two-dimensional (2D) separable transform. Scaling factors and clip bit widths determined for a particular transpose buffer size and the expected transform sizes are used to reduce the size of the intermediate results of applying the 2D separable transform. The reduced bit widths of the intermediate results may vary across the intermediate results. In some embodiments, the scaling factors and associated clip bit widths may be adapted during encoding.