Neural Inverse Rendering for General Reflectance Photometric Stereo

TL;DR

This paper introduces a physics-based unsupervised neural network approach for photometric stereo that accurately recovers surface normals and reflectance properties without requiring ground truth data, achieving state-of-the-art results.

Contribution

It presents a novel unsupervised learning framework that predicts surface normals and BRDFs directly from images, overcoming data acquisition challenges and invariance issues.

Findings

Achieves state-of-the-art performance on real-world benchmarks.

Does not require ground truth normals or pre-training.

Successfully models complex reflectance properties.

Abstract

We present a novel convolutional neural network architecture for photometric stereo (Woodham, 1980), a problem of recovering 3D object surface normals from multiple images observed under varying illuminations. Despite its long history in computer vision, the problem still shows fundamental challenges for surfaces with unknown general reflectance properties (BRDFs). Leveraging deep neural networks to learn complicated reflectance models is promising, but studies in this direction are very limited due to difficulties in acquiring accurate ground truth for training and also in designing networks invariant to permutation of input images. In order to address these challenges, we propose a physics based unsupervised learning framework where surface normals and BRDFs are predicted by the network and fed into the rendering equation to synthesize observed images. The network weights are optimized during testing by minimizing reconstruction loss between observed and synthesized images. Thus, our learning process does not require ground truth normals or even pre-training on external images. Our method is shown to achieve the state-of-the-art performance on a challenging real-world scene benchmark.