Robust Photometric Stereo via Dictionary Learning

Andrew J. Wagenmaker, Brian E. Moore, Raj Rao Nadakuditi

Published 2017-10-24Version 1

Photometric stereo is a method that seeks to reconstruct the normal vectors of an object from a set of images of the object illuminated under different light sources. While effective in some situations, classical photometric stereo relies on a diffuses surface model that cannot handle objects with complex reflectance patterns, and it is sensitive to non-idealities in the images. In this work, we propose a novel approach to photometric stereo that relies on dictionary learning to produce robust normal vector reconstructions. Specifically, we develop three formulations for applying dictionary learning to photometric stereo. We propose a preprocessing step that utilizes dictionary learning to denoise the images. We also present a model that applies dictionary learning to regularize and reconstruct the normal vectors from the images under the classic Lambertian reflectance model. Finally, we generalize the latter model to explicitly model non-Lambertian objects. We investigate all three approaches through extensive experimentation on synthetic and real benchmark datasets and observe state-of-the-art performance compared to existing robust photometric stereo methods.