Texture compression and synthesis method with fine granularity and high compression rate

Abstract

The invention provides a texture compression and synthesis method with fine granularity and high compression rate, wherein the compression method comprises the following steps: generating an MIPMAP with low resolution; dividing the MIPMAP with low resolution into a plurality of basic compression units; generating basis color data and modulation data on each basic compression unit; and forming all texture compression blocks into texture compression data. The synthesis method comprises the following steps: carrying out linear interpolation on basic color and adjacent basic color data in each compression block to generate temporary intermediate basic color data; taking out the modulation data; modulating the intermediate basic color data to construct basic units; combining all generated basic units to construct the MIPMAP with low resolution; and carrying out bilinear interpolation on the generated MIPMAP to construct a target texture map with high resolution. The invention has the advantages that different modulation modes are suitable for different occasions and the compression quality is excellent; and the modulation data are directly saved in the compression block.

Description

technical field [0001] The invention relates to a texture compression and synthesis method in computer graphics, in particular to a fine-grained and high compression rate texture compression and synthesis method. Background technique [0002] Computer Graphics (CG) is a science that uses mathematical algorithms to convert two-dimensional or three-dimensional graphics into a raster form for computer displays. Simply put, the main research content of computer graphics is to study how to represent graphics in the computer, and to use the computer to calculate, process and display the related principles and algorithms of graphics. [0003] A texture is an array of one-, two-, or multi-dimensional data items used in computing the color or appearance of a fragment resulting from the rasterization of a computer graphics picture. Textures can be image data, color or transparency data, roughness / smoothness data, albedo data, etc. Delivering realistic computer graphics generally req...

AI Technical Summary

Problems solved by technology

[0006] 2. The BD method of data block decomposition is limited by the color types in the color table, and cannot handle the problems of "sharp texture boundaries" or "multiple texture colors" well, and the quality of reconstruction is poor;

[0007] 3. Both DCT-based compression method and low-resolution low-pass filter image method require too complicated texture decompression process, which is difficult or expensive to implement

Images