Cross-Component Adaptive Loop Filter Coefficient Constraint
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The dynamic range of intermedia data during the filtering process in cross-component adaptive loop filters (CC-ALF) for video coding exceeds 16-bit, which is undesirable, especially for software implementation, and existing solutions do not effectively constrain the range of CC-ALF filter coefficients.
Innovation Solution
Constraining the CC-ALF filter coefficients to be less than the sum of the bit-depth value associated with the luma samples and a positive offset value, ensuring the dynamic range remains within a preferred limit, such as 16-bit, by using a cross-component adaptive loop filter that extracts luma samples to refine the chroma component and applies linear diamond-shaped filters with scaled and rounded coefficients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If CC-ALF filter coefficients are used to refine chroma component based on luma samples, then filtering quality is improved, but dynamic range of intermedia data exceeds 16-bit which complicates software implementation
Solution Approach 1:
The patent constrains the CC-ALF filter coefficients to specific ranges based on bit-depth parameters. By changing the parameter constraints of filter coefficients to be within representable ranges of standard bit-depth formats (e.g., 8-bit, 10-bit, 12-bit), the patent ensures that filtering quality is maintained while the dynamic range of intermedia data remains within 16-bit limits, making software implementation feasible without requiring arbitrary precision arithmetic.
2Ease of manufacture
If filter coefficients are constrained to maintain 16-bit dynamic range, then software implementation becomes feasible, but filtering effectiveness may be reduced
Solution Approach 1:
The patent establishes specific parameter ranges for filter coefficients based on bit-depth values. By carefully selecting coefficient ranges that are functions of the input signal bit-depth (e.g., coefficients constrained to be within [-2^(bit-depth-1), 2^(bit-depth-1)]), the patent achieves optimal filtering effectiveness while ensuring that intermediate calculations do not exceed 16-bit dynamic range, thus maintaining both software feasibility and filtering precision.
Solution Approach 2:
The patent applies partial action by using only the necessary portion of the full dynamic range for coefficient representation. Instead of allowing coefficients to use the complete theoretical dynamic range, the patent constrains them to a subset that is sufficient for effective filtering but guarantees that when combined with input samples, the result remains within 16-bit representable range, avoiding overflow while maintaining filtering effectiveness.
Data Source
AI summary
A method, computer program, and computer system is provided for video coding. Video data comprising a chroma component and a luma component is received. Luma samples are extracted from the luma component of the received video data. The chroma component is filtered based on the extracted luma samples using a cross-component adaptive loop filter (CC-ALF). Coefficients associated with the CC-ALF are constrained to be less than a sum of a bit-depth value associated with the luma samples and a positive offset value.