Neural Texture Block Compression

TL;DR

This paper introduces Neural Texture Block Compression (NTBC), a neural network-based method that reduces texture storage size by up to 70% while maintaining real-time performance in graphics applications.

Contribution

NTBC is a novel neural network approach that learns to compress textures efficiently without altering shader operations, improving storage efficiency over traditional fixed-ratio methods.

Findings

Achieves up to 70% reduction in texture storage footprint.

Maintains real-time performance with modest computational overhead.

Provides reasonable-quality texture compression results.

Abstract

Block compression is a widely used technique to compress textures in real-time graphics applications, offering a reduction in storage size. However, their storage efficiency is constrained by the fixed compression ratio, which substantially increases storage size when hundreds of high-quality textures are required. In this paper, we propose a novel block texture compression method with neural networks, Neural Texture Block Compression (NTBC). NTBC learns the mapping from uncompressed textures to block-compressed textures, which allows for significantly reduced storage costs without any change in the shaders.Our experiments show that NTBC can achieve reasonable-quality results with up to about 70% less storage footprint, preserving real-time performance with a modest computational overhead at the texture loading phase in the graphics pipeline.