Inferring geometry and material properties from Mueller matrices with machine learning
This work addresses a domain-specific problem in optics and material science for researchers and engineers, showing incremental progress by applying machine learning to a known bottleneck.
The researchers tackled the ill-posed problem of simultaneously recovering geometry and material properties from Mueller matrices by using machine learning on a dataset of spheres with isotropic materials, demonstrating that surface normals can be predicted and geometry reconstructed even with unknown material types, and material types can be correctly identified, with diagonal elements key for material characterization and off-diagonal elements decisive for normal estimation.
Mueller matrices (MMs) encode information on geometry and material properties, but recovering both simultaneously is an ill-posed problem. We explore whether MMs contain sufficient information to infer surface geometry and material properties with machine learning. We use a dataset of spheres of various isotropic materials, with MMs captured over the full angular domain at five visible wavelengths (450-650 nm). We train machine learning models to predict material properties and surface normals using only these MMs as input. We demonstrate that, even when the material type is unknown, surface normals can be predicted and object geometry reconstructed. Moreover, MMs allow models to identify material types correctly. Further analyses show that diagonal elements are key for material characterization, and off-diagonal elements are decisive for normal estimation.