SD-MVS: Segmentation-Driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization

TL;DR

SD-MVS introduces a segmentation-driven approach using SAM and EM optimization to improve 3D reconstruction quality, especially in textureless areas, achieving state-of-the-art results efficiently.

Contribution

The paper presents a novel segmentation-driven multi-view stereo method that integrates SAM for semantic constraints and EM for parameter optimization, enhancing reconstruction completeness and robustness.

Findings

Achieves state-of-the-art results on ETH3D and Tanks and Temples datasets.

Effectively handles textureless regions in 3D reconstruction.

Reduces time consumption compared to existing methods.

Abstract

In this paper, we introduce Segmentation-Driven Deformation Multi-View Stereo (SD-MVS), a method that can effectively tackle challenges in 3D reconstruction of textureless areas. We are the first to adopt the Segment Anything Model (SAM) to distinguish semantic instances in scenes and further leverage these constraints for pixelwise patch deformation on both matching cost and propagation. Concurrently, we propose a unique refinement strategy that combines spherical coordinates and gradient descent on normals and pixelwise search interval on depths, significantly improving the completeness of reconstructed 3D model. Furthermore, we adopt the Expectation-Maximization (EM) algorithm to alternately optimize the aggregate matching cost and hyperparameters, effectively mitigating the problem of parameters being excessively dependent on empirical tuning. Evaluations on the ETH3D high-resolution multi-view stereo benchmark and the Tanks and Temples dataset demonstrate that our method can achieve state-of-the-art results with less time consumption.