{"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/exploring-object-centric-temporal-modeling","title":"Exploring Object-Centric Temporal Modeling for Efficient Multi-View 3D Object Detection","arxiv_id":"2303.11926","date":"2023-03-21","proceeding":"ICCV 2023 1","authors":["Shihao Wang","Yingfei Liu","Tiancai Wang","Ying Li","Xiangyu Zhang"],"abstract":"In this paper, we propose a long-sequence modeling framework, named StreamPETR, for multi-view 3D object detection. Built upon the sparse query design in the PETR series, we systematically develop an object-centric temporal mechanism. The model is performed in an online manner and the long-term historical information is propagated through object queries frame by frame. Besides, we introduce a motion-aware layer normalization to model the movement of the objects. StreamPETR achieves significant performance improvements only with negligible computation cost, compared to the single-frame baseline. On the standard nuScenes benchmark, it is the first online multi-view method that achieves comparable performance (67.6% NDS & 65.3% AMOTA) with lidar-based methods. The lightweight version realizes 45.0% mAP and 31.7 FPS, outperforming the state-of-the-art method (SOLOFusion) by 2.3% mAP and 1.8x faster FPS. Code has been available at https://github.com/exiawsh/StreamPETR.git.","url_abs":"https://arxiv.org/abs/2303.11926v2","url_pdf":"https://arxiv.org/pdf/2303.11926v2.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":"exploring-object-centric-temporal-modeling","repo_url":"https://github.com/exiawsh/streampetr","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"exploring-object-centric-temporal-modeling","repo_url":"https://github.com/wenyuqing/panacea","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"3d-multi-object-tracking","task_name":"3D Multi-Object Tracking"},{"task_slug":"3d-object-detection","task_name":"3D Object Detection"},{"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":"layer-normalization","method_name":"Layer Normalization"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-multi-object-tracking-on-nuscenes-camera-1","task":"3D Multi-Object Tracking","dataset":"nuScenes Camera Only","model":"StreamPETR-Large","rank_in_archive_order":1,"of":1,"metrics":{"AMOTA":"65.3"},"uses_additional_data":false},{"leaderboard":"/sota/3d-object-detection-on-3d-object-detection-on","task":"3D Object Detection","dataset":"3D Object Detection on Argoverse2 Camera Only","model":"StreamPETR","rank_in_archive_order":2,"of":3,"metrics":{"Average mAP":"20.3"},"uses_additional_data":false},{"leaderboard":"/sota/3d-object-detection-on-nuscenes-camera-only","task":"3D Object Detection","dataset":"nuScenes Camera Only","model":"StreamPETR-Large","rank_in_archive_order":4,"of":19,"metrics":{"Future Frame":"false","NDS":"67.6"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2303.11926","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}