Generative Modelling of BRDF Textures from Flash Images
This addresses the challenge for graphics and vision researchers of efficiently modeling material appearance from limited data, though it is incremental as it builds on existing latent space and CNN techniques.
The paper tackles the problem of capturing and reproducing visual material appearance from a single flash photograph by learning a latent space for material codes and generating diverse spatial fields of BRDF parameters, enabling rendering in complex scenes that matches the input. A user study shows favorable comparison to previous methods, including those with BRDF supervision.
We learn a latent space for easy capture, consistent interpolation, and efficient reproduction of visual material appearance. When users provide a photo of a stationary natural material captured under flashlight illumination, first it is converted into a latent material code. Then, in the second step, conditioned on the material code, our method produces an infinite and diverse spatial field of BRDF model parameters (diffuse albedo, normals, roughness, specular albedo) that subsequently allows rendering in complex scenes and illuminations, matching the appearance of the input photograph. Technically, we jointly embed all flash images into a latent space using a convolutional encoder, and -- conditioned on these latent codes -- convert random spatial fields into fields of BRDF parameters using a convolutional neural network (CNN). We condition these BRDF parameters to match the visual characteristics (statistics and spectra of visual features) of the input under matching light. A user study compares our approach favorably to previous work, even those with access to BRDF supervision.