TexNN: Fast Texture Encoding Using Neural Networks

• Published in: Computer graphics forum Vol. 38; no. 1; pp. 328 - 339
• Main Authors: Pratapa, S., Olson, T., Chalfin, A., Manocha, D.
• Format: Journal Article
• Language: English
• Published: Oxford Blackwell Publishing Ltd 01.02.2019
• Subjects: Algorithms; CCS Concepts •Computing methodologies→ Neural networks; Image compression; Supervised learning; Graphics file formats; Coding; Compressing; Compression tests; Computation; Computer graphics; Configurations; data compression; GPUs and their application for general‐purpose computing; hardware; image and video processing; image compression; Machine learning; modelling; Neural networks; Texture
• ISSN: 0167-7055; 1467-8659
• DOI: 10.1111/cgf.13534

We present a novel deep learning‐based method for fast encoding of textures into current texture compression formats. Our approach uses state‐of‐the‐art neural network methods to compute the appropriate encoding configurations for fast compression. A key bottleneck in the current encoding algorithms is the search step, and we reduce that computation to a classification problem. We use a trained neural network approximation to quickly compute the encoding configuration for a given texture. We have evaluated our approach for compressing the textures for the widely used adaptive scalable texture compression format and evaluate the performance for different block sizes corresponding to 4 × 4, 6 × 6 and 8 × 8. Overall, our method (TexNN) speeds up the encoding computation up to an order of magnitude compared to prior compression algorithms with very little or no loss in the visual quality. We present a novel deep learning‐based method for fast encoding of textures into current texture compression formats. Our approach uses state‐of‐the‐art neural network methods to compute the appropriate encoding configurations for fast compression. A key bottleneck in the current encoding algorithms is the search step, and we reduce that computation to a classification problem. We use a trained neural network approximation to quickly compute the encoding configuration for a given texture. We have evaluated our approach for compressing the textures for the widely used adaptive scalable texture compression format and evaluate the performance for different block sizes corresponding to 4 × 4, 6 × 6 and 8 × 8.