{"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/cagroup3d-class-aware-grouping-for-3d-object","title":"CAGroup3D: Class-Aware Grouping for 3D Object Detection on Point Clouds","arxiv_id":"2210.04264","date":"2022-10-09","proceeding":null,"authors":["Haiyang Wang","Lihe Ding","Shaocong Dong","Shaoshuai Shi","Aoxue Li","Jianan Li","Zhenguo Li","LiWei Wang"],"abstract":"We present a novel two-stage fully sparse convolutional 3D object detection framework, named CAGroup3D. Our proposed method first generates some high-quality 3D proposals by leveraging the class-aware local group strategy on the object surface voxels with the same semantic predictions, which considers semantic consistency and diverse locality abandoned in previous bottom-up approaches. Then, to recover the features of missed voxels due to incorrect voxel-wise segmentation, we build a fully sparse convolutional RoI pooling module to directly aggregate fine-grained spatial information from backbone for further proposal refinement. It is memory-and-computation efficient and can better encode the geometry-specific features of each 3D proposal. Our model achieves state-of-the-art 3D detection performance with remarkable gains of +\\textit{3.6\\%} on ScanNet V2 and +\\textit{2.6}\\% on SUN RGB-D in term of mAP@0.25. Code will be available at https://github.com/Haiyang-W/CAGroup3D.","url_abs":"https://arxiv.org/abs/2210.04264v1","url_pdf":"https://arxiv.org/pdf/2210.04264v1.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":[{"paper_slug":"cagroup3d-class-aware-grouping-for-3d-object","repo_url":"https://github.com/haiyang-w/cagroup3d","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"3d-object-detection","task_name":"3D Object Detection"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-object-detection-on-sun-rgbd","task":"3D Object Detection","dataset":"SUN-RGBD","model":"CAGroup3D (Geo Only)","rank_in_archive_order":2,"of":8,"metrics":{"mAP@0.25":"66.8","mAP@0.5":"50.2"},"uses_additional_data":false},{"leaderboard":"/sota/3d-object-detection-on-sun-rgbd-val","task":"3D Object Detection","dataset":"SUN-RGBD val","model":"CAGroup3D(Geo only)","rank_in_archive_order":8,"of":32,"metrics":{"mAP@0.25":"66.8","mAP@0.5":"50.2"},"uses_additional_data":false},{"leaderboard":"/sota/3d-object-detection-on-scannetv2","task":"3D Object Detection","dataset":"ScanNetV2","model":"CAGroup3D","rank_in_archive_order":9,"of":33,"metrics":{"mAP@0.25":"75.1","mAP@0.5":"61.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2210.04264","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}