Image Gradient-Aided Photometric Stereo Network

TL;DR

This paper introduces IGA-PSN, a dual-branch neural network that leverages both photometric images and their gradients to improve surface normal estimation in photometric stereo, especially around high-frequency details.

Contribution

The paper proposes a novel dual-branch framework with gradient information and an hourglass regression network, enhancing normal estimation accuracy over existing methods.

Findings

Outperforms previous methods on DiLiGenT benchmarks

Achieves a mean angular error of 6.46 degrees

Preserves textures and geometric details in complex regions

Abstract

Photometric stereo (PS) endeavors to ascertain surface normals using shading clues from photometric images under various illuminations. Recent deep learning-based PS methods often overlook the complexity of object surfaces. These neural network models, which exclusively rely on photometric images for training, often produce blurred results in high-frequency regions characterized by local discontinuities, such as wrinkles and edges with significant gradient changes. To address this, we propose the Image Gradient-Aided Photometric Stereo Network (IGA-PSN), a dual-branch framework extracting features from both photometric images and their gradients. Furthermore, we incorporate an hourglass regression network along with supervision to regularize normal regression. Experiments on DiLiGenT benchmarks show that IGA-PSN outperforms previous methods in surface normal estimation, achieving a mean angular error of 6.46 while preserving textures and geometric shapes in complex regions.