{"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/monocular-quasi-dense-3d-object-tracking","title":"Monocular Quasi-Dense 3D Object Tracking","arxiv_id":"2103.07351","date":"2021-03-12","proceeding":null,"authors":["Hou-Ning Hu","Yung-Hsu Yang","Tobias Fischer","Trevor Darrell","Fisher Yu","Min Sun"],"abstract":"A reliable and accurate 3D tracking framework is essential for predicting future locations of surrounding objects and planning the observer's actions in numerous applications such as autonomous driving. We propose a framework that can effectively associate moving objects over time and estimate their full 3D bounding box information from a sequence of 2D images captured on a moving platform. The object association leverages quasi-dense similarity learning to identify objects in various poses and viewpoints with appearance cues only. After initial 2D association, we further utilize 3D bounding boxes depth-ordering heuristics for robust instance association and motion-based 3D trajectory prediction for re-identification of occluded vehicles. In the end, an LSTM-based object velocity learning module aggregates the long-term trajectory information for more accurate motion extrapolation. Experiments on our proposed simulation data and real-world benchmarks, including KITTI, nuScenes, and Waymo datasets, show that our tracking framework offers robust object association and tracking on urban-driving scenarios. On the Waymo Open benchmark, we establish the first camera-only baseline in the 3D tracking and 3D detection challenges. Our quasi-dense 3D tracking pipeline achieves impressive improvements on the nuScenes 3D tracking benchmark with near five times tracking accuracy of the best vision-only submission among all published methods. Our code, data and trained models are available at https://github.com/SysCV/qd-3dt.","url_abs":"https://arxiv.org/abs/2103.07351v1","url_pdf":"https://arxiv.org/pdf/2103.07351v1.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":"monocular-quasi-dense-3d-object-tracking","repo_url":"https://github.com/SysCV/qd-3dt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-object-tracking","task_name":"3D Object Tracking"},{"task_slug":"autonomous-driving","task_name":"Autonomous Driving"},{"task_slug":"multiple-object-tracking","task_name":"Multiple Object Tracking"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"trajectory-prediction","task_name":"Trajectory Prediction"}],"methods":[],"datasets_introduced":[{"slug":"3d-vehicle-tracking-simulation-dataset","name":"3D Vehicle Tracking Simulation Dataset","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/multiple-object-tracking-on-kitti-test-online","task":"Multiple Object Tracking","dataset":"KITTI Test (Online Methods)","model":"QD-3DT","rank_in_archive_order":16,"of":34,"metrics":{"HOTA":"72.77 ","MOTA":"86.41"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2103.07351","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}