Revisiting Photometric Ambiguity for Accurate Gaussian-Splatting Surface Reconstruction

TL;DR

AmbiSuR introduces a novel framework that addresses photometric ambiguities in Gaussian Splatting for more accurate 3D surface reconstruction, leveraging intrinsic ambiguity self-indication and disambiguation techniques.

Contribution

The paper uncovers primitive ambiguities in Gaussian Splatting and proposes a disambiguation and self-indication module to improve surface reconstruction accuracy.

Findings

Outperforms existing methods in various challenging scenarios.

Demonstrates superior surface reconstruction quality and robustness.

Achieves broad compatibility across different datasets.

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/ .