Deep SVBRDF Estimation on Real Materials
arXiv:2010.04143 · doi:10.1109/3DV50981.2020.00126
Abstract
Recent work has demonstrated that deep learning approaches can successfully be used to recover accurate estimates of the spatially-varying BRDF (SVBRDF) of a surface from as little as a single image. Closer inspection reveals, however, that most approaches in the literature are trained purely on synthetic data, which, while diverse and realistic, is often not representative of the richness of the real world. In this paper, we show that training such networks exclusively on synthetic data is insufficient to achieve adequate results when tested on real data. Our analysis leverages a new dataset of real materials obtained with a novel portable multi-light capture apparatus. Through an extensive series of experiments and with the use of a novel deep learning architecture, we explore two strategies for improving results on real data: finetuning, and a per-material optimization procedure. We show that adapting network weights to real data is of critical importance, resulting in an approach which significantly outperforms previous methods for SVBRDF estimation on real materials. Dataset and code are available at https://lvsn.github.io/real-svbrdf
arXiv commentAccepted submission to 3DV 2020. Project page https://lvsn.github.io/real-svbrdf
References in corpus (5)
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Single-Image SVBRDF Capture with a Rendering-Aware Deep Network
- Modeling Surface Appearance from a Single Photograph using Self-augmented Convolutional Neural Networks
- Two-shot Spatially-varying BRDF and Shape Estimation
- Guided Fine-Tuning for Large-Scale Material Transfer