PointMC: Multi-instance Point Cloud Registration based on Maximal Cliques

Yue Wu, Xidao Hu, Yongzhe Yuan, Xiaolong Fan, Maoguo Gong, Hao Li, Mingyang Zhang, Qiguang Miao, Wenping Ma
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:53542-53552, 2024.

Abstract

Multi-instance point cloud registration is the problem of estimating multiple rigid transformations between two point clouds. Existing solutions rely on global spatial consistency of ambiguity and the time-consuming clustering of highdimensional correspondence features, making it difficult to handle registration scenarios where multiple instances overlap. To address these problems, we propose a maximal clique based multiinstance point cloud registration framework called PointMC. The key idea is to search for maximal cliques on the correspondence compatibility graph to estimate multiple transformations, and cluster these transformations into clusters corresponding to different instances to efficiently and accurately estimate all poses. PointMC leverages a correspondence embedding module that relies on local spatial consistency to effectively eliminate outliers, and the extracted discriminative features empower the network to circumvent missed pose detection in scenarios involving multiple overlapping instances. We conduct comprehensive experiments on both synthetic and real-world datasets, and the results show that the proposed PointMC yields remarkable performance improvements.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-wu24k, title = {{P}oint{MC}: Multi-instance Point Cloud Registration based on Maximal Cliques}, author = {Wu, Yue and Hu, Xidao and Yuan, Yongzhe and Fan, Xiaolong and Gong, Maoguo and Li, Hao and Zhang, Mingyang and Miao, Qiguang and Ma, Wenping}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {53542--53552}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/wu24k/wu24k.pdf}, url = {https://proceedings.mlr.press/v235/wu24k.html}, abstract = {Multi-instance point cloud registration is the problem of estimating multiple rigid transformations between two point clouds. Existing solutions rely on global spatial consistency of ambiguity and the time-consuming clustering of highdimensional correspondence features, making it difficult to handle registration scenarios where multiple instances overlap. To address these problems, we propose a maximal clique based multiinstance point cloud registration framework called PointMC. The key idea is to search for maximal cliques on the correspondence compatibility graph to estimate multiple transformations, and cluster these transformations into clusters corresponding to different instances to efficiently and accurately estimate all poses. PointMC leverages a correspondence embedding module that relies on local spatial consistency to effectively eliminate outliers, and the extracted discriminative features empower the network to circumvent missed pose detection in scenarios involving multiple overlapping instances. We conduct comprehensive experiments on both synthetic and real-world datasets, and the results show that the proposed PointMC yields remarkable performance improvements.} }
Endnote
%0 Conference Paper %T PointMC: Multi-instance Point Cloud Registration based on Maximal Cliques %A Yue Wu %A Xidao Hu %A Yongzhe Yuan %A Xiaolong Fan %A Maoguo Gong %A Hao Li %A Mingyang Zhang %A Qiguang Miao %A Wenping Ma %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-wu24k %I PMLR %P 53542--53552 %U https://proceedings.mlr.press/v235/wu24k.html %V 235 %X Multi-instance point cloud registration is the problem of estimating multiple rigid transformations between two point clouds. Existing solutions rely on global spatial consistency of ambiguity and the time-consuming clustering of highdimensional correspondence features, making it difficult to handle registration scenarios where multiple instances overlap. To address these problems, we propose a maximal clique based multiinstance point cloud registration framework called PointMC. The key idea is to search for maximal cliques on the correspondence compatibility graph to estimate multiple transformations, and cluster these transformations into clusters corresponding to different instances to efficiently and accurately estimate all poses. PointMC leverages a correspondence embedding module that relies on local spatial consistency to effectively eliminate outliers, and the extracted discriminative features empower the network to circumvent missed pose detection in scenarios involving multiple overlapping instances. We conduct comprehensive experiments on both synthetic and real-world datasets, and the results show that the proposed PointMC yields remarkable performance improvements.
APA
Wu, Y., Hu, X., Yuan, Y., Fan, X., Gong, M., Li, H., Zhang, M., Miao, Q. & Ma, W.. (2024). PointMC: Multi-instance Point Cloud Registration based on Maximal Cliques. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:53542-53552 Available from https://proceedings.mlr.press/v235/wu24k.html.

Related Material