{"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/do-you-remember-the-future-weak-to-strong","title":"Do You Remember . . . the Future? Weak-to-Strong generalization in 3D Object Detection","arxiv_id":null,"date":"2024-08-03","proceeding":"IJCAI 2024 8","authors":["Alexander Gambashidze","Aleksandr Dadukin","Maxim Golyadkin","Maria Razzhivina","Ilya Makarov"],"abstract":"This paper demonstrates a novel method for\r\nLiDAR-based 3D object detection, addressing ma-\r\njor field challenges: sparsity and occlusion. Our\r\napproach leverages temporal point cloud sequences\r\nto generate frames that provide comprehensive\r\nviews of objects from multiple angles. To address\r\nthe challenge of generating these frames in real-\r\ntime, we employ Knowledge Distillation within\r\na Teacher-Student framework, allowing the Stu-\r\ndent model to emulate the Teacher’s advanced per-\r\nception. We pioneered the application of weak-\r\nto-strong generalization in computer vision by\r\ntraining our Teacher model on enriched, object-\r\ncomplete data. In this demo, we showcase the ex-\r\nceptional quality of labels produced by the X-Ray\r\nTeacher on object-complete frames, showing our\r\nmethod distilling its knowledge to enhance object\r\n3D detection models.","url_abs":"https://www.ijcai.org/proceedings/2024/1001","url_pdf":"https://www.ijcai.org/proceedings/2024/1001.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":"do-you-remember-the-future-weak-to-strong","repo_url":"https://github.com/sakharok13/x-ray-teacher-patching-tools","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"3d-object-detection","task_name":"3D Object Detection"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-object-detection-on-waymo-open-dataset","task":"3D Object Detection","dataset":"Waymo Open Dataset","model":"X-Ray DSVT Pillar-Scaled","rank_in_archive_order":5,"of":8,"metrics":{"mAPH/L2":"71.4"},"uses_additional_data":false},{"leaderboard":"/sota/3d-object-detection-on-nuscenes","task":"3D Object Detection","dataset":"nuScenes","model":"X-Ray CenterPoint-Voxel","rank_in_archive_order":151,"of":372,"metrics":{"NDS":"0.63","mAP":"0.54"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}