Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.24431 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908716211634176 |
|---|---|
| author | Zheng, Bozhong Gan, Jinye Xu, Xiaohao Chen, Xintao Li, Wenqiao Huang, Xiaonan Ni, Na Wu, Yingna |
| author_facet | Zheng, Bozhong Gan, Jinye Xu, Xiaohao Chen, Xintao Li, Wenqiao Huang, Xiaonan Ni, Na Wu, Yingna |
| contents | 3D point cloud anomaly detection is essential for robust vision systems but is challenged by pose variations and complex geometric anomalies. Existing patch-based methods often suffer from geometric fidelity issues due to discrete voxelization or projection-based representations, limiting fine-grained anomaly localization. We introduce Pose-Aware Signed Distance Field (PASDF), a novel framework that integrates 3D anomaly detection and repair by learning a continuous, pose-invariant shape representation. PASDF leverages a Pose Alignment Module for canonicalization and a SDF Network to dynamically incorporate pose, enabling implicit learning of high-fidelity anomaly repair templates from the continuous SDF. This facilitates precise pixel-level anomaly localization through an Anomaly-Aware Scoring Module. Crucially, the continuous 3D representation in PASDF extends beyond detection, facilitating in-situ anomaly repair. Experiments on Real3D-AD and Anomaly-ShapeNet demonstrate state-of-the-art performance, achieving high object-level AUROC scores of 80.2% and 90.0%, respectively. These results highlight the effectiveness of continuous geometric representations in advancing 3D anomaly detection and facilitating practical anomaly region repair. The code is available at https://github.com/ZZZBBBZZZ/PASDF to support further research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_24431 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Bridging 3D Anomaly Localization and Repair via High-Quality Continuous Geometric Representation Zheng, Bozhong Gan, Jinye Xu, Xiaohao Chen, Xintao Li, Wenqiao Huang, Xiaonan Ni, Na Wu, Yingna Computer Vision and Pattern Recognition 3D point cloud anomaly detection is essential for robust vision systems but is challenged by pose variations and complex geometric anomalies. Existing patch-based methods often suffer from geometric fidelity issues due to discrete voxelization or projection-based representations, limiting fine-grained anomaly localization. We introduce Pose-Aware Signed Distance Field (PASDF), a novel framework that integrates 3D anomaly detection and repair by learning a continuous, pose-invariant shape representation. PASDF leverages a Pose Alignment Module for canonicalization and a SDF Network to dynamically incorporate pose, enabling implicit learning of high-fidelity anomaly repair templates from the continuous SDF. This facilitates precise pixel-level anomaly localization through an Anomaly-Aware Scoring Module. Crucially, the continuous 3D representation in PASDF extends beyond detection, facilitating in-situ anomaly repair. Experiments on Real3D-AD and Anomaly-ShapeNet demonstrate state-of-the-art performance, achieving high object-level AUROC scores of 80.2% and 90.0%, respectively. These results highlight the effectiveness of continuous geometric representations in advancing 3D anomaly detection and facilitating practical anomaly region repair. The code is available at https://github.com/ZZZBBBZZZ/PASDF to support further research. |
| title | Bridging 3D Anomaly Localization and Repair via High-Quality Continuous Geometric Representation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.24431 |