Select-and-Combine (SAC): A Novel Multi-Stereo Depth Fusion Algorithm for Point Cloud Generation via Efficient Local Markov Netlets

TL;DR

This paper introduces SAC, a novel multi-stereo depth fusion algorithm that selects the best depth per point using local Markov Netlets, resulting in more accurate and cleaner point clouds for surface reconstruction.

Contribution

The paper proposes a new depth fusion paradigm using local Markov Netlets for point selection and combination, addressing non-Gaussian errors and improving accuracy and clarity.

Findings

Outperforms existing methods with a 2.07% higher F1 score.

Produces point clouds that are 18% less redundant.

Achieves clearer and more accurate surface reconstructions.

Abstract

Many practical systems for image-based surface reconstruction employ a stereo/multi-stereo paradigm, due to its ability to scale for large scenes and its ease of implementation for out-of-core operations. In this process, multiple and abundant depth maps from stereo matching must be combined and fused into a single, consistent, and clean point cloud. However, the noises and outliers caused by stereo matching and the heterogenous geometric errors of the poses present a challenge for existing fusion algorithms, since they mostly assume Gaussian errors and predict fused results based on data from local spatial neighborhoods, which may inherit uncertainties from multiple depths resulting in lowered accuracy. In this paper, we propose a novel depth fusion paradigm, that instead of numerically fusing points from multiple depth maps, selects the best depth map per point, and combines them into a single and clean point cloud. This paradigm, called select-and-combine (SAC), is achieved through modeling the point level fusion using local Markov Netlets, a micro-network over point across neighboring views for depth/view selection, followed by a Netlets collapse process for point combination. The Markov Netlets are optimized such that they can inherently leverage spatial consistencies among depth maps of neighboring views, thus they can address errors beyond Gaussian ones. Our experiment results show that our approach outperforms existing depth fusion approaches by increasing the F1 score that considers both accuracy and completeness by 2.07% compared to the best existing method. Finally, our approach generates clearer point clouds that are 18% less redundant while with a higher accuracy before fusion