Deep Polarization Cues for Single-shot Shape and Subsurface Scattering Estimation

TL;DR

This paper introduces a learning-based approach that leverages polarization cues to jointly estimate shape and subsurface scattering parameters of translucent objects, supported by a new synthetic dataset and outperforming existing methods.

Contribution

It is the first to utilize polarization cues for subsurface scattering estimation and provides a large-scale synthetic dataset for training and evaluation.

Findings

Outperforms baseline methods on synthetic data

Effective in real-world scenarios

Utilizes polarization cues for improved SSS estimation

Abstract

In this work, we propose a novel learning-based method to jointly estimate the shape and subsurface scattering (SSS) parameters of translucent objects by utilizing polarization cues. Although polarization cues have been used in various applications, such as shape from polarization (SfP), BRDF estimation, and reflection removal, their application in SSS estimation has not yet been explored. Our observations indicate that the SSS affects not only the light intensity but also the polarization signal. Hence, the polarization signal can provide additional cues for SSS estimation. We also introduce the first large-scale synthetic dataset of polarized translucent objects for training our model. Our method outperforms several baselines from the SfP and inverse rendering realms on both synthetic and real data, as demonstrated by qualitative and quantitative results.