Abstract:Three-dimensional point cloud registration is a critical task in 3D perception for sensors that aims to determine the optimal alignment between two point clouds by finding the best transformation. Existing methods like RANSAC and its variants often face challenges, such as sensitivity to low overlap rates, high computational costs, and susceptibility to outliers, leading to inaccurate results, especially in complex or noisy environments. In this paper, we introduce a novel 3D registration method, CL-PCR, inspired by the concept of maximal cliques and built upon the SC2-PCR framework. Our approach allows for the flexible use of smaller sampling subsets to extract more local consensus information, thereby generating accurate pose hypotheses even in scenarios with low overlap between point clouds. This method enhances robustness against low overlap and reduces the influence of outliers, addressing the limitations of traditional techniques. First, we construct a graph matrix to represent the compatibility relationships among the initial correspondences. Next, we build clique-likes subsets of various sizes within the graph matrix, each representing a consensus set. Then, we compute the transformation hypotheses for the subsets using the SVD algorithm and select the best hypothesis for registration based on evaluation metrics. Extensive experiments demonstrate the effectiveness of CL-PCR. In comparison experiments on the 3DMatch/3DLoMatch datasets using both FPFH and FCGF descriptors, our Fast-CL-PCRv1 outperforms state-of-the-art algorithms, achieving superior registration performance. Additionally, we validate the practicality and robustness of our method with real-world data.

A Clustering Successive POCS Algorithm for Fast Point Matching

A Clustering And Quadratic Programming Based Pocs Algorithm For Point Matching

A Quadratic Programming Based Cluster Correspondence Projection Algorithm for Fast Point Matching

Cluster analysis via projection onto convex sets

TOCSAC: TOpology Constraint SAmple Consensus for Fast and Reliable Feature Correspondence

Feature Embedding Clustering using POCS-based Clustering Algorithm

A Discriminative Point Matching Algorithm Based on Local Structure Consensus Constraint

A Practical Fast Point Pattern Matching Algorithm Based On Matching Pair Support And Registration Condition

A Robust Algorithm For Feature Point Matching

A Fast Point Cloud Recognition Algorithm Based on Keypoint Pair Feature

An Efficient Smoothing Proximal Gradient Algorithm for Convex Clustering

Clique-like Point Cloud Registration: A Flexible Sampling Registration Method Based on Clique-like for Low-Overlapping Point Cloud

Outlier robust point correspondence based on GNCCP.

A Polyhedral Annexation Algorithm for Aligning Partially Overlapping Point Sets

Point Pattern Matching for Articulated or Multiple Objects

Multi-instance Point Cloud Registration by Efficient Correspondence Clustering

3D Point Pattern Matching Based on Spatial Geometric Flexibility.

Dilated POCS: Minimax Convex Optimization

An image matching optimization algorithm based on pixel shift clustering RANSAC

A Point Pattern Matching Algorithm Based on Probabilistic Relaxation

A Simple and Robust Feature Point Matching Algorithm Based on Restricted Spatial Order Constraints for Aerial Image Registration