Revisiting Photometric Ambiguity for Accurate Gaussian-Splatting Surface Reconstruction

2026-05-12

AI Summary

This work addresses the pervasive photometric ambiguity in differentiable rendering–based surface reconstruction by leveraging Gaussian splatting representations. It is the first to identify two distinct primitive-level photometric ambiguities inherent in this framework and introduces a self-indicating mechanism to resolve them. Specifically, an ambiguity indication module detects under-constrained regions, which are then refined through a photometric disambiguation constraint. This approach significantly enhances the determinacy and accuracy of geometric reconstruction, outperforming existing methods across a variety of complex scenes while achieving high-fidelity and highly compatible surface recovery.

Abstract

Surface reconstruction with differentiable rendering has achieved impressive performance in recent years, yet the pervasive photometric ambiguities have strictly bottlenecked existing approaches. This paper presents AmbiSuR, a framework that explores an intrinsic solution upon Gaussian Splatting for the photometric ambiguity-robust surface 3D reconstruction with high performance. Starting by revisiting the foundation, our investigation uncovers two built-in primitive-wise ambiguities in representation, while revealing an intrinsic potential for ambiguity self-indication in Gaussian Splatting. Stemming from these, a photometric disambiguation is first introduced, constraining ill-posed geometry solution for definite surface formation. Then, we propose an ambiguity indication module that unleashes the self-indication potential to identify and further guide correcting underconstrained reconstructions. Extensive experiments demonstrate our superior surface reconstructions compared to existing methods across various challenging scenarios, excelling in broad compatibility. Project: https://fictionarry.github.io/AmbiSuR-Proj/ .

Problem

• photometric ambiguity
• surface reconstruction
• Gaussian Splatting
• 3D reconstruction
• differentiable rendering

Innovation

• photometric ambiguity
• Gaussian Splatting
• surface reconstruction
• differentiable rendering
• ambiguity self-indication