Image Gradient-Aided Photometric Stereo Network
This addresses surface normal estimation for photometric stereo, offering incremental improvements in handling complex object surfaces.
The paper tackled the problem of blurred results in high-frequency regions of photometric stereo by proposing a dual-branch network that uses both photometric images and their gradients, achieving a mean angular error of 6.46 on benchmarks.
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.