SinPoint: A Novel Topological Consistent Augmentation for Point Cloud Understanding

Jian Bi, Qianliang Wu, Xiang Li, Shuo Chen, Jianjun Qian, lei luo, Jian Yang

Keywords: Topological consistent, homeomorphism, Point clouds augmentation

TL;DR: A simple and effective point cloud augmentation method based on homeomorphism

Abstract: Data augmentation is a highly effective method for addressing the issue of data scarcity in machine learning and computer vision tasks. It involves diversifying the original data through a series of transformations to improve the robustness and generalization ability of the model. However, due to the disorder and irregularity of point clouds, existing methods struggle to enrich geometric diversity and maintain topological consistency, leading to imprecise point cloud understanding. In this paper, we propose SinPoint, a novel method designed to preserve the topological structure of the original point cloud through a homeomorphism. Additionally, it utilizes the Sine function to generate smooth displacements. This simulates object deformations, thereby producing a rich diversity of samples. Our extensive experiments demonstrate that SinPoint consistently outperforms existing Mixup and Deformation methods on various benchmark point cloud datasets, improving performance for shape classification and part segmentation tasks. Specifically, when used with PointNet++ and DGCNN, SinPoint achieves a state-of-the-art accuracy of 90.2 on shape classification with the real-world ScanObjectNN dataset. Furthermore, our method is highly versatile and scalable, and it can adapt to different scenarios and requirements for point cloud tasks.

Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 5358