Neural BRDFs: Representation and Operations
This work addresses the need for flexible and efficient BRDF manipulation in computer graphics for realistic appearance rendering, representing an incremental improvement over existing neural BRDF methods by adding operational capabilities.
The paper tackles the problem of representing and manipulating bidirectional reflectance distribution functions (BRDFs) in computer graphics by introducing a neural network-based approach that compresses BRDFs into latent vectors and enables operations like layering and interpolation directly in the latent space, resulting in accurate representation and efficient evaluation and sampling that competes with more expensive Monte Carlo methods.
Bidirectional reflectance distribution functions (BRDFs) are pervasively used in computer graphics to produce realistic physically-based appearance. In recent years, several works explored using neural networks to represent BRDFs, taking advantage of neural networks' high compression rate and their ability to fit highly complex functions. However, once represented, the BRDFs will be fixed and therefore lack flexibility to take part in follow-up operations. In this paper, we present a form of "Neural BRDF algebra", and focus on both representation and operations of BRDFs at the same time. We propose a representation neural network to compress BRDFs into latent vectors, which is able to represent BRDFs accurately. We further propose several operations that can be applied solely in the latent space, such as layering and interpolation. Spatial variation is straightforward to achieve by using textures of latent vectors. Furthermore, our representation can be efficiently evaluated and sampled, providing a competitive solution to more expensive Monte Carlo layering approaches.