Real-time Neural Woven Fabric Rendering
This enables real-time, high-quality rendering and editing of woven fabrics for applications like computer graphics and virtual reality, though it is incremental as it builds on neural methods for material representation.
The paper tackles real-time rendering of woven fabrics by proposing a lightweight neural network that encodes fabric patterns and parameters into a small latent vector, achieving nearly 60 FPS on an RTX 3090 with quality close to ground truth and no visible aliasing or noise.
Woven fabrics are widely used in applications of realistic rendering, where real-time capability is also essential. However, rendering realistic woven fabrics in real time is challenging due to their complex structure and optical appearance, which cause aliasing and noise without many samples. The core of this issue is a multi-scale representation of the fabric shading model, which allows for a fast range query. Some previous neural methods deal with the issue at the cost of training on each material, which limits their practicality. In this paper, we propose a lightweight neural network to represent different types of woven fabrics at different scales. Thanks to the regularity and repetitiveness of woven fabric patterns, our network can encode fabric patterns and parameters as a small latent vector, which is later interpreted by a small decoder, enabling the representation of different types of fabrics. By applying the pixel's footprint as input, our network achieves multi-scale representation. Moreover, our network is fast and occupies little storage because of its lightweight structure. As a result, our method achieves rendering and editing woven fabrics at nearly 60 frames per second on an RTX 3090, showing a quality close to the ground truth and being free from visible aliasing and noise.