Deep Polarization Imaging for 3D shape and SVBRDF Acquisition

TL;DR

This paper introduces a deep learning-based method that uses polarization imaging to efficiently capture 3D shape and surface reflectance of objects from a single view, surpassing previous techniques in quality.

Contribution

It combines polarization cues with deep learning to estimate shape and SVBRDF from a single polarization image without prior shape constraints.

Findings

Achieves high-quality 3D shape and reflectance estimation from single-view polarization images.

Outperforms recent deep learning methods with flash illumination in accuracy.

Provides qualitative improvements through network architecture and loss modifications.

Abstract

We present a novel method for efficient acquisition of shape and spatially varying reflectance of 3D objects using polarization cues. Unlike previous works that have exploited polarization to estimate material or object appearance under certain constraints (known shape or multiview acquisition), we lift such restrictions by coupling polarization imaging with deep learning to achieve high quality estimate of 3D object shape (surface normals and depth) and SVBRDF using single-view polarization imaging under frontal flash illumination. In addition to acquired polarization images, we provide our deep network with strong novel cues related to shape and reflectance, in the form of a normalized Stokes map and an estimate of diffuse color. We additionally describe modifications to network architecture and training loss which provide further qualitative improvements. We demonstrate our approach to achieve superior results compared to recent works employing deep learning in conjunction with flash illumination.