Local Illumination Compensation Using SAD Linear Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video encoding and decoding methods face inefficiencies in Local Illumination Compensation (LIC) processes due to high computational complexity, leading to potential visual artifacts and reduced coding efficiency.
Innovation Solution
The implementation of a method that uses a linear model for LIC, where parameters are computed using sums of absolute differences of neighboring reconstructed and reference samples, with a regularization process to reduce complexity and prevent discontinuity issues, thereby improving the estimation of local illumination changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional LIC parameter computation methods are used, then estimation accuracy of local illumination changes is improved, but computational complexity increases
Solution Approach 1:
The patent changes the computational parameters by using sums of absolute differences (SAD) instead of more complex optimization algorithms. The linear model parameters (a, b) are derived using simplified equations based on SAD computations between reference blocks and current blocks, maintaining estimation accuracy while significantly reducing computational complexity
Solution Approach 2:
The patent employs computationally inexpensive SAD operations that can be quickly computed and discarded, replacing more expensive and time-consuming optimization algorithms. The temporary variables and intermediate computations used in the simplified derivation are lightweight and do not require substantial computational resources
2Measurement precision
If conventional LIC parameter computation methods are used, then estimation accuracy of local illumination changes is improved, but visual artifacts increase
Solution Approach 1:
The patent incorporates a feedback mechanism where the linear model parameters are iteratively refined using SAD computations. The derivation process continuously adjusts parameters (a, b) based on the comparison between reference block predictions and actual current block values, reducing estimation errors and resulting visual artifacts
Solution Approach 2:
By using SAD-based parameter derivation instead of conventional methods, the patent achieves more stable and accurate illumination change estimates. The simplified mathematical derivation leads to parameters that better represent actual illumination variations, thereby reducing artifacts in the reconstructed video
3Device complexity
If simplified LIC parameter computation is used, then computational complexity is reduced, but estimation accuracy of local illumination changes deteriorates
Solution Approach 1:
The patent replaces complex mechanical-like optimization processes with a streamlined mathematical derivation based on SAD. Instead of using iterative optimization algorithms that require multiple computational passes, the patent employs closed-form solutions derived from SAD computations, achieving both simplicity and accuracy
Data Source
AI summary
Described are methods and apparatuses for encoding/decoding a picture comprising predicting at least one block, wherein the predicting comprises performing motion compensation and local illumination compensation based on a reference block, the local illumination compensation including applying a linear model based on sums of absolute differences of neighboring reconstructed samples and corresponding reference samples of the reference block, wherein the neighboring reconstructed samples and corresponding reference samples of the reference block are co-located according to an L-shape substantially adjacent to the block to be predicted, the L-shape comprising a row of pixels located to the top side of the predicted block and a column of pixels located to the left side of the predicted block, the co-location being determined according a motion vector of the predicted block.