{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/taming-anomalies-with-down-up-sampling","title":"Taming Anomalies with Down-Up Sampling Networks: Group Center Preserving Reconstruction for 3D Anomaly Detection","arxiv_id":"2507.03903","date":"2025-07-05","proceeding":null,"authors":["Hanzhe Liang","Jie Zhang","Tao Dai","Linlin Shen","Jinbao Wang","Can Gao"],"abstract":"Reconstruction-based methods have demonstrated very promising results for 3D anomaly detection. However, these methods face great challenges in handling high-precision point clouds due to the large scale and complex structure. In this study, a Down-Up Sampling Network (DUS-Net) is proposed to reconstruct high-precision point clouds for 3D anomaly detection by preserving the group center geometric structure. The DUS-Net first introduces a Noise Generation module to generate noisy patches, which facilitates the diversity of training data and strengthens the feature representation for reconstruction. Then, a Down-sampling Network~(Down-Net) is developed to learn an anomaly-free center point cloud from patches with noise injection. Subsequently, an Up-sampling Network (Up-Net) is designed to reconstruct high-precision point clouds by fusing multi-scale up-sampling features. Our method leverages group centers for construction, enabling the preservation of geometric structure and providing a more precise point cloud. Extensive experiments demonstrate the effectiveness of our proposed method, achieving state-of-the-art (SOTA) performance with an Object-level AUROC of 79.9% and 79.5%, and a Point-level AUROC of 71.2% and 84.7% on the Real3D-AD and Anomaly-ShapeNet datasets, respectively.","url_abs":"https://arxiv.org/abs/2507.03903v1","url_pdf":"https://arxiv.org/pdf/2507.03903v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"3d-anomaly-detection","task_name":"3D Anomaly Detection"},{"task_slug":"anomaly-detection","task_name":"Anomaly Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-anomaly-detection-on-anomaly-shapenet","task":"3D Anomaly Detection","dataset":"Anomaly-ShapeNet","model":"DUS-Net","rank_in_archive_order":5,"of":8,"metrics":{"O-AUROC":"0.797","P-AUROC":"0.712"},"uses_additional_data":false},{"leaderboard":"/sota/3d-anomaly-detection-on-real-3d-ad","task":"3D Anomaly Detection","dataset":"Real 3D-AD","model":"DUS-Net","rank_in_archive_order":1,"of":19,"metrics":{"Mean Performance of P. and O. ":"0.821","Object AUROC":"0.795","Point AUROC":"0.847"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2507.03903","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}