Incorporating Lambertian Priors Into Surface Normals Measurement

Abstract

The goal of photometric stereo is to measure the precise surface normal of a 3-D object from observations with various shading cues. However, non-Lambertian surfaces influence the measurement accuracy due to irregular shading cues. Despite deep neural networks being used to simulate the performance of non-Lambertian surfaces, the error in specularities, shadows, and crinkle regions is hard to be reduced. To address this challenge, we here propose a photometric stereo network that incorporates Lambertian priors to better measure the surface normal. In this article, we use the initial normal under the Lambertian assumption as prior information to refine the normal measurement, instead of solely applying the observed shading cues to deriving the surface normal. Our method uses the Lambertian information to reparameterize the network weights and the powerful fitting ability of deep neural networks to correct these errors caused by general reflectance properties. Our explorations include: the Lambertian priors: 1) reduce the learning hypothesis space, making our method learn mapping in the same surface normal space and improving the accuracy of learning and 2) provides the differential features' learning, improving the surfaces' reconstruction of details. Extensive experiments verify the effectiveness of the proposed Lambertian prior photometric stereo network in accurate surface normal measurement, on the challenging benchmark dataset.