TY - JOUR A1 - Henzler, Philipp A1 - Deschaintre, Valentin A1 - Mitra, Niloy J A1 - Ritschel, Tobias JF - ACM Transactions on Graphics UR - https://doi.org/10.1145/3478513.3480507 SN - 0730-0301 IS - 6 N1 - This version is the author accepted manuscript. For information on re-use, please refer to the publisher?s terms and conditions. VL - 40 KW - Science & Technology KW - Technology KW - Computer Science KW - Software Engineering KW - Computer Science KW - material capture KW - appearance capture KW - SVBRDF KW - deep learning KW - generative model KW - unsupervised learning KW - ILLUMINATION KW - REFLECTANCE PB - ASSOC COMPUTING MACHINERY N2 - 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. Project webpage: https://henzler.github.io/publication/neuralmaterial/. ID - discovery10159070 AV - public Y1 - 2021/12/01/ EP - 13 TI - Generative Modelling of BRDF Textures from Flash Images ER -