Video Coding Filter Using Frequency Domain Gain Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in maintaining video quality while reducing bandwidth requirements, often requiring complex rate-distortion optimization and large numbers of weighting coefficients for efficient processing, which increases complexity and reduces coding efficiency.
Innovation Solution
A filter that processes video blocks by loading pixels into a linear buffer, performing 1D transforms, multiplying spectrum components with gain coefficients derived from quantization parameters, and generating filtered blocks, which reduces the need for complex signaling and universal multipliers, allowing for efficient video coding with low complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If adaptive loop filter with weighting coefficients is used to improve video quality, then video quality is improved, but device complexity increases due to complex rate-distortion optimization and large numbers of weighting coefficients
Solution Approach 1:
The patent changes the filtering approach from spatial domain weighted sums to frequency domain operations using 1D transforms. By transforming pixels to spectrum components, applying simple gain coefficients (derived from quantization parameters rather than complex weighting coefficients), and transforming back, the method achieves similar quality improvement with significantly reduced complexity and fewer parameters to manage.
Solution Approach 2:
The patent replaces the mechanical computation of complex weighted sums in the spatial domain with a substitution approach using frequency domain transforms. Instead of directly computing weighted sums of neighboring pixels, the method uses 1D transforms to convert to frequency domain, applies simplified gain coefficients, and transforms back, thereby substituting a complex mechanical filtering process with a more efficient transform-based approach.
2Manufacturing precision
If adaptive loop filter with many weighting coefficients is used to improve video quality, then video quality is improved, but bandwidth requirements increase due to transmission of filter parameters
Solution Approach 1:
The patent fundamentally changes the filter parameters from complex weighting coefficients (which require extensive transmission and storage) to simple gain coefficients derived directly from quantization parameters. This parameter transformation dramatically reduces the amount of data that needs to be transmitted and stored while maintaining effective filtering capability through the frequency domain approach.
3Manufacturing precision
If universal multipliers are used in adaptive loop filter to improve video quality, then video quality is improved, but device complexity increases due to multiplier reloading for each pixel block
Solution Approach 1:
The patent substitutes the mechanical multiplier-based filtering system with a transform-based system using 1D transforms and gain coefficient multiplication. By converting to frequency domain, applying simple gain factors (which can be implemented with shift operations rather than full multipliers), and transforming back, the method eliminates the need for complex universal multipliers that must be reloaded for each 2x2 pixel block, significantly reducing hardware complexity.
Data Source
AI summary
A filter for video coding is provided, where the filter is configured for processing a block for generation of a filtered block, and the block comprises a plurality of pixels. The filter includes one or more processor configured to: obtain a quantization parameter (QP) of the block; obtain a threshold (THR) based on the QP; and obtain a look up table based on QP, so as to generate a filtered block based on the threshold and the look up table. The filter is provided allowing improving the efficiency for video coding.