High quality image processing

Abstract

In a high quality image-encoding scheme an input image is decomposed into several image blocks comprising multiple image elements. The image blocks are encoded into encoded block representations. Such a block representation comprises two color codewords, a color modifying codeword and optionally a sequence of color indices and color modifier indices. The color codewords define multiple discrete color representations along a line in color space. The color modifying codeword represents a set of multiple color modifiers for modifying the multiple color representations along at least one extension vector to obtain, for each color representation, a set of multiple color points. These color points of the multiple sets are located on a surface defined by the multiple color representations and the at least one extension vector. The colors of the image elements in the block are then approximated by these color points on the surface.

Description

TECHNICAL FIELD[0001]The present invention generally refers to image processing, and in particular to methods and systems for encoding and decoding images at a high image quality.BACKGROUND[0002]Presentation and rendering of images and graphics on data processing systems and user terminals, such as computers, and in particular on mobile terminals have increased tremendously the last years. For example, three-dimensional (3D) graphics and images have a number of appealing applications on such terminals, including games, 3D maps and messaging, screen savers and man-machine interfaces.[0003]A 3D graphics rendering process typically comprises three sub-stages. Briefly, a first stage, the application stage, creates several triangles. The corners of these triangles are transformed, projected and lit in a second stage, the geometry stage. In a third stage, the rasterization stage, images, often denoted textures, can be “glued” onto the triangles, increasing the realism of the rendered imag...

AI Technical Summary

Benefits of technology

The present invention provides a high-quality image processing and compression scheme that can be used in existing compression methods. The invention involves decomposing an image into multiple blocks, each containing multiple image elements, and using a color modifying code to modify the color representations of these elements along an extension vector. This results in a surface in color space that is defined by the modified color representations and the direction vector. The invention also provides a method for approximating the colors of image elements using multiple color points on a surface in color space. The use of multiple color modifiers and a direction vector allows for a more accurate representation of the colors of the image. The invention can be used in both lossy and lossless compression methods.

Problems solved by technology

However, a memory lookup in the palette is required, and the palette is restricted in size.

One disadvantage of S3TC is that only four colors can be used per block.

The major problem in terms of quality with this method is that the chrominance is heavily quantized, which may introduce block artifacts.

Despite the relative high quality obtainable by PACKMAN, iPACKMAN and S3TC, this quality is still not enough for some applications.

However, one of the problems of running the user interface on the 3D hardware is the limited memory bandwidth.

Furthermore, it is expected that the quality of today's texture compression schemes may not be good enough for compression of user interface constants and textures.

For example, icons and other small graphical symbols are usually very colorful and of high contrast, which does not work well with most compression schemes.

The are also applications, such as games on personal computers and game consoles, where compression of some textures with the most widely used scheme, S3TC, does not give enough quality.

Images