{"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/point-cloud-pre-training-with-natural-3d","title":"Point Cloud Pre-Training With Natural 3D Structures","arxiv_id":null,"date":"2022-01-01","proceeding":"CVPR 2022 1","authors":["Ryosuke Yamada","Hirokatsu Kataoka","Naoya Chiba","Yukiyasu Domae","Tetsuya OGATA"],"abstract":"    The construction of 3D point cloud datasets requires a great deal of human effort. Therefore, constructing a largescale 3D point clouds dataset is difficult. In order to remedy this issue, we propose a newly developed point cloud fractal database (PC-FractalDB), which is a novel family of formula-driven supervised learning inspired by fractal geometry encountered in natural 3D structures. Our research is based on the hypothesis that we could learn representations from more real-world 3D patterns than conventional 3D datasets by learning fractal geometry. We show how the PC-FractalDB facilitates solving several recent dataset-related problems in 3D scene understanding, such as 3D model collection and labor-intensive annotation. The experimental section shows how we achieved the performance rate of up to 61.9% and 59.0% for the ScanNetV2 and SUN RGB-D datasets, respectively, over the current highest scores obtained with the PointContrast, contrastive scene contexts (CSC), and RandomRooms. Moreover, the PC-FractalDB pre-trained model is especially effective in training with limited data. For example, in 10% of training data on ScanNetV2, the PC-FractalDB pre-trained VoteNet performs at 38.3%, which is +14.8% higher accuracy than CSC. Of particular note, we found that the proposed method achieves the highest results for 3D object detection pre-training in limited point cloud data.    ","url_abs":"http://openaccess.thecvf.com//content/CVPR2022/html/Yamada_Point_Cloud_Pre-Training_With_Natural_3D_Structures_CVPR_2022_paper.html","url_pdf":"http://openaccess.thecvf.com//content/CVPR2022/papers/Yamada_Point_Cloud_Pre-Training_With_Natural_3D_Structures_CVPR_2022_paper.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":"point-cloud-pre-training-with-natural-3d","repo_url":"https://github.com/ryosuke-yamada/3dfractaldb","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"3d-object-detection","task_name":"3D Object Detection"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"point-cloud-pre-training","task_name":"Point Cloud Pre-training"},{"task_slug":"scene-understanding","task_name":"Scene Understanding"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-object-detection-on-sun-rgbd-val","task":"3D Object Detection","dataset":"SUN-RGBD val","model":"VoteNet (PC-FractalDB)","rank_in_archive_order":20,"of":32,"metrics":{"mAP@0.25":"60.2","mAP@0.5":"35.2"},"uses_additional_data":true},{"leaderboard":"/sota/3d-object-detection-on-scannetv2","task":"3D Object Detection","dataset":"ScanNetV2","model":"VoteNet (PC-FractalDB)","rank_in_archive_order":25,"of":33,"metrics":{"mAP@0.25":"63.4","mAP@0.5":"39.9"},"uses_additional_data":true}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}