PO3AD: Predicting Point Offsets toward Better 3D Point Cloud Anomaly Detection

Jianan Ye, Weiguang Zhao, Xi Yang, Guangliang Cheng, Kaizhu Huang

Keywords: 3D point cloud, anomaly detection

Abstract: Point cloud anomaly detection, particularly under the anomaly-free setting, poses a significant challenge as it requires the precise capture of 3D normal data features to accurately identify deviations indicative of anomalies. Current efforts focus on devising reconstruction tasks, such as acquiring normal data representations by restoring normal samples from altered, pseudo-anomalous counterparts. Nonetheless, such methods tend to dilute the model's focus, as they require attention to both normal and pseudo-anomalous data points, thereby hampering the efficacy of the learning process. Moreover, the inherently disordered and sparse nature of 3D point cloud data significantly complicates the task. In response to those predicaments, we introduce an innovative approach that involves learning point offsets for the first time, with a concentrated emphasis on more informative pseudo-abnormal points, thus fostering more effective distillation of normal data representations. We have crafted an augmentation technique that is steered by normal vectors, facilitating the creation of credible pseudo anomalies that enhance the efficiency of the training process. Our comprehensive experimental evaluation on the Anomaly-ShapeNet and Real3D-AD datasets evidences that our proposed method outperforms existing state-of-the-art approaches, achieving an average enhancement of 9.0% and 1.4% in the AUC-ROC detection metric across these datasets, respectively.

Primary Area: applications to computer vision, audio, language, and other modalities

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: 5835