Segmentation-Driven Monocular Shape from Polarization based on Physical Model

TL;DR

This paper presents a segmentation-driven framework for monocular shape-from-polarization that improves surface normal estimation by locally segmenting the surface and applying adaptive convexity priors, reducing azimuth ambiguity.

Contribution

It introduces a novel segmentation strategy and multi-scale convexity prior to enhance shape recovery accuracy in polarization-based 3D reconstruction.

Findings

Significant reduction in azimuth ambiguity.

Improved surface normal accuracy over existing methods.

Enhanced detail recovery in real-world datasets.

Abstract

Monocular shape-from-polarization (SfP) leverages the intrinsic relationship between light polarization properties and surface geometry to recover surface normals from single-view polarized images, providing a compact and robust approach for three-dimensional (3D) reconstruction. Despite its potential, existing monocular SfP methods suffer from azimuth angle ambiguity, an inherent limitation of polarization analysis, that severely compromises reconstruction accuracy and stability. This paper introduces a novel segmentation-driven monocular SfP (SMSfP) framework that reformulates global shape recovery into a set of local reconstructions over adaptively segmented convex sub-regions. Specifically, a polarization-aided adaptive region growing (PARG) segmentation strategy is proposed to decompose the global convexity assumption into locally convex regions, effectively suppressing azimuth ambiguities and preserving surface continuity. Furthermore, a multi-scale fusion convexity prior (MFCP) constraint is developed to ensure local surface consistency and enhance the recovery of fine textural and structural details. Extensive experiments on both synthetic and real-world datasets validate the proposed approach, showing significant improvements in disambiguation accuracy and geometric fidelity compared with existing physics-based monocular SfP techniques.