Low Complexity Decimated Grid Selection in ASTC Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Advanced Scalable Texture Compression (ASTC) encoding systems require complex iterative approaches for block mode selection, making them unsuitable for real-time encoding applications, particularly in power-constrained devices like smartphones and tablets.
Innovation Solution
A low-complexity method for selecting an optimal decimated grid is introduced, which estimates weight decimation errors for vertical and horizontal grids, combines these with color format choice errors, and selects the grids with the minimum combined errors as optimal, thereby reducing computational complexity and improving encoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional iterative block mode selection is used in ASTC encoding, then encoding quality is optimized, but computational complexity increases making it unsuitable for real-time applications
Solution Approach 1:
The patent segments the complex iterative block mode selection process into distinct functional modules: a complexity reduction module that pre-processes and filters candidate modes, and a selection module that chooses from the reduced set. This segmentation eliminates unnecessary iterations while preserving encoding quality, making real-time ASTC encoding feasible on power-constrained devices.
2Manufacturing precision
If conventional iterative block mode selection is used, then optimal compression is achieved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by performing complexity reduction before the actual mode selection process. The complexity reduction module pre-calculates and filters candidate block modes based on preliminary error metrics, so that the subsequent selection process works with a reduced set of promising candidates. This preliminary filtering dramatically reduces processing time while maintaining optimal compression efficiency.
3Measurement precision
If full precision block mode selection is performed, then encoding accuracy is maximized, but power consumption increases
Solution Approach 1:
The patent applies partial action by performing full precision error calculation only for a reduced set of candidate block modes identified by the complexity reduction module, rather than evaluating all possible modes with full precision. This partial application of computationally intensive operations maintains encoding accuracy for the most promising candidates while significantly reducing overall power consumption.
Data Source
AI summary
A method, executed by an encoding circuit, for selecting an optimal decimated grid includes estimating, among a plurality of decimated grids within a coded block, a weight decimation error for each of a vertical decimated grid and a horizontal decimated grid. A weight decimation error is estimated for each of remaining decimated grids among the plurality of decimated grids based on a corresponding weight decimation error among the vertical decimated grid and the horizontal decimated grid. For each of the plurality of decimated grids, an initial combined error is determined by combining the weight decimation error and a color format choice error. Two decimated grids having the least initial combined errors, among the plurality of decimated grids, are selected as optimal decimated grids.