Progressive Correspondence Regenerator for Robust 3D Registration

2025-02-04

AI Summary

Traditional correspondence optimization methods fail in 3D registration under extreme outlier ratios (e.g., >95%), struggling to simultaneously maximize the number and accuracy of correct matches. To address this, we propose Regor, a progressive correspondence regeneration framework. Departing from mainstream outlier rejection paradigms, Regor introduces the novel concept of “correspondence regeneration”: it first generates initial correspondences via prior-guided local grouping and generalized mutual matching, then refines them locally using center-aware triplet consistency constraints, and finally performs global refinement across multiple iterative stages. Its core innovation lies in modeling robust registration as a continuous process of high-quality correspondence generation—not a one-shot outlier removal. Evaluated on diverse indoor and outdoor benchmarks, Regor achieves up to 10× more inliers than state-of-the-art outlier rejection methods, significantly improving registration robustness and accuracy under challenging conditions such as weak texture and low overlap.

Technology Category

• Search and Optimization: Learning to Search
• Intelligent Robots: Learning & Optimization for ROB
• Computer Vision: 3D Computer Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Obtaining enough high-quality correspondences is crucial for robust registration. Existing correspondence refinement methods mostly follow the paradigm of outlier removal, which either fails to correctly identify the accurate correspondences under extreme outlier ratios, or select too few correct correspondences to support robust registration. To address this challenge, we propose a novel approach named Regor, which is a progressive correspondence regenerator that generates higher-quality matches whist sufficiently robust for numerous outliers. In each iteration, we first apply prior-guided local grouping and generalized mutual matching to generate the local region correspondences. A powerful center-aware three-point consistency is then presented to achieve local correspondence correction, instead of removal. Further, we employ global correspondence refinement to obtain accurate correspondences from a global perspective. Through progressive iterations, this process yields a large number of high-quality correspondences. Extensive experiments on both indoor and outdoor datasets demonstrate that the proposed Regor significantly outperforms existing outlier removal techniques. More critically, our approach obtain 10 times more correct correspondences than outlier removal methods. As a result, our method is able to achieve robust registration even with weak features. The code will be released.

Problem

• Improves 3D registration robustness
• Generates high-quality correspondences
• Handles extreme outlier ratios effectively

Innovation

• Progressive correspondence regenerator
• Center-aware three-point consistency
• Global correspondence refinement