Partial Coverage Layers for Lossless Image Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional graphics processors face challenges in compressing image data with high contrast between pixels and samples within a tile, often resulting in high data loss during compression.
Innovation Solution
The implementation of partial coverage layers that encode image data without reading existing data stored in the tile, allowing for efficient compression and reduced bandwidth usage by generating and writing these layers without accessing existing image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional compression methods are used on tiles with high contrast graphics primitives, then compression ratio is improved, but data loss increases significantly
Solution Approach 1:
The tile is divided into multiple sub-tiles based on coverage patterns. Each sub-tile contains pixels with similar coverage characteristics, allowing different compression strategies to be applied to each segment. This segmentation enables lossless compression for high-contrast regions while maintaining overall compression efficiency.
Solution Approach 2:
Different compression quality levels are applied to different regions within the tile based on their specific characteristics. Regions with high contrast and partial coverage use lossless compression methods, while uniform regions use lossy compression. This local quality approach ensures that data loss only occurs in regions where it is imperceptible.
2Manufacturing precision
If existing image data is read before writing compressed data to a tile, then compression accuracy is improved, but bandwidth usage increases
Solution Approach 1:
Coverage patterns and compression parameters are determined in advance based on primitive geometry calculations before actual pixel data is generated or accessed. This preliminary analysis of coverage patterns allows the system to prepare compression strategies without reading existing tile data, eliminating the need for data reads while maintaining compression accuracy.
Solution Approach 2:
The compression system uses only the newly generated image data and coverage information to perform compression, without requiring access to existing tile data. The algorithm is self-sufficient in that it can determine all necessary compression parameters from the input primitives and coverage patterns alone, eliminating bandwidth-consuming reads of existing data.
3Device complexity
If full coverage layers are used for all pixels in a tile, then compression simplicity is improved, but memory footprint increases when tiles are not fully covered
Solution Approach 1:
Instead of applying full coverage layer compression to all pixels in a tile, the system applies compression only to the specific sub-tiles that are actually covered by graphics primitives. This partial action approach processes only the necessary portions of the tile, reducing memory footprint while maintaining compression simplicity for the covered regions.
Data Source
AI summary
One embodiment of the present invention sets forth a technique for compressing image data with high contrast between pixels within a tile and between samples within pixels without any data loss. Partial coverage layers are generated and written to a tile that includes multiple pixels without reading the existing image data that is stored for the tile. A partial coverage layer encodes image data, such as colors, and sub-pixel coverage information for each covered pixel in a tile. The use of partial coverage layers reduces the bandwidth used to store image data when a tile is not fully covered.