Salvato in:
Dettagli Bibliografici
Autori principali: Lian, Wei, Cui, Zhesen, Ma, Fei, Pan, Hang, Zuo, Wangmeng, Zhang, Jianmei
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2405.08589
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915537223680000
author Lian, Wei
Cui, Zhesen
Ma, Fei
Pan, Hang
Zuo, Wangmeng
Zhang, Jianmei
author_facet Lian, Wei
Cui, Zhesen
Ma, Fei
Pan, Hang
Zuo, Wangmeng
Zhang, Jianmei
contents Numerous applications require algorithms that can align partially overlapping point sets while maintaining invariance to geometric transformations (e.g., similarity, affine, rigid). This paper introduces a novel global optimization method for this task by minimizing the objective function of the Robust Point Matching (RPM) algorithm. We first reveal that the original RPM objective is a cubic polynomial. Through a concise variable substitution, we transform this objective into a quadratic function. By leveraging the convex envelope of bilinear monomials, we derive a tight lower bound for this quadratic function. This lower bound problem conveniently and efficiently decomposes into two parts: a standard linear assignment problem (solvable in polynomial time) and a low-dimensional convex quadratic program. Furthermore, we devise a specialized Branch-and-Bound (BnB) algorithm that branches exclusively on the transformation parameters, which significantly accelerates convergence by confining the search space. Experiments on 2D and 3D synthetic and real-world data demonstrate that our method, compared to state-of-the-art approaches, exhibits superior robustness to non-rigid deformations, positional noise, and outliers, particularly in scenarios where outliers are distinct from inliers.
format Preprint
id arxiv_https___arxiv_org_abs_2405_08589
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Decomposed Global Optimization for Robust Point Matching with Low-Dimensional Branching
Lian, Wei
Cui, Zhesen
Ma, Fei
Pan, Hang
Zuo, Wangmeng
Zhang, Jianmei
Computer Vision and Pattern Recognition
Numerous applications require algorithms that can align partially overlapping point sets while maintaining invariance to geometric transformations (e.g., similarity, affine, rigid). This paper introduces a novel global optimization method for this task by minimizing the objective function of the Robust Point Matching (RPM) algorithm. We first reveal that the original RPM objective is a cubic polynomial. Through a concise variable substitution, we transform this objective into a quadratic function. By leveraging the convex envelope of bilinear monomials, we derive a tight lower bound for this quadratic function. This lower bound problem conveniently and efficiently decomposes into two parts: a standard linear assignment problem (solvable in polynomial time) and a low-dimensional convex quadratic program. Furthermore, we devise a specialized Branch-and-Bound (BnB) algorithm that branches exclusively on the transformation parameters, which significantly accelerates convergence by confining the search space. Experiments on 2D and 3D synthetic and real-world data demonstrate that our method, compared to state-of-the-art approaches, exhibits superior robustness to non-rigid deformations, positional noise, and outliers, particularly in scenarios where outliers are distinct from inliers.
title Decomposed Global Optimization for Robust Point Matching with Low-Dimensional Branching
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.08589