JCSE, vol. 10, no. 2, pp.68-73, June, 2016

DOI: Block matching; Texture compression; Spatial similarity

A Hybrid Texture Coding Method for Fast Texture Mapping

Li Cui, Hyungyu Kim, and Euee S. Jang
Department of Computer & Software, Hanyang University, Seoul, Korea and Department of Computer Science & Technology, College of Science & Engineering, Yanbian University, Yanji, China Department of Electronics & Computer Engineering, Hanyang University, Seoul, Korea Division of Computer Science & Engineering, Hanyang University, Seoul, Korea

Abstract: An efficient texture compression method is proposed based on a block matching process between the current block and the previously encoded blocks. Texture mapping is widely used to improve the quality of rendering results in real-time applications. For fast texture mapping, it is important to find an optimal trade-off between compression efficiency and computational complexity. Low-complexity methods (e.g., ETC1 and DXT1) have often been adopted in real-time rendering applications because conventional compression methods (e.g., JPEG) achieve a high compression ratio at the cost of high complexity. We propose a block matching-based compression method that can achieve a higher compression ratio than ETC1 and DXT1 while maintaining computational complexity lower than that of JPEG. Through a comparison between the proposed method and existing compression methods, we confirm our expectations on the performance of the proposed method.

Keyword: Block matching; Texture compression; Spatial similarity

Full Paper:   307 Downloads, 1187 View