Learning a Neural 3D Texture Space from 2D Exemplars
This work addresses texture generation for computer graphics applications, offering a novel method that enables 3D texture learning from 2D inputs without per-exemplar retraining.
The paper tackles the problem of generating diverse and high-fidelity 2D and 3D natural textures from 2D exemplars, achieving computational efficiency by extending stochastic procedural texturing with learned deep non-linearities.
We propose a generative model of 2D and 3D natural textures with diversity, visual fidelity and at high computational efficiency. This is enabled by a family of methods that extend ideas from classic stochastic procedural texturing (Perlin noise) to learned, deep, non-linearities. The key idea is a hard-coded, tunable and differentiable step that feeds multiple transformed random 2D or 3D fields into an MLP that can be sampled over infinite domains. Our model encodes all exemplars from a diverse set of textures without a need to be re-trained for each exemplar. Applications include texture interpolation, and learning 3D textures from 2D exemplars.