{"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/fusion-of-head-and-full-body-detectors-for","title":"Fusion of Head and Full-Body Detectors for Multi-Object Tracking","arxiv_id":"1705.08314","date":"2017-05-23","proceeding":null,"authors":["Roberto Henschel","Laura Leal-Taixé","Daniel Cremers","Bodo Rosenhahn"],"abstract":"In order to track all persons in a scene, the tracking-by-detection paradigm\nhas proven to be a very effective approach. Yet, relying solely on a single\ndetector is also a major limitation, as useful image information might be\nignored. Consequently, this work demonstrates how to fuse two detectors into a\ntracking system. To obtain the trajectories, we propose to formulate tracking\nas a weighted graph labeling problem, resulting in a binary quadratic program.\nAs such problems are NP-hard, the solution can only be approximated. Based on\nthe Frank-Wolfe algorithm, we present a new solver that is crucial to handle\nsuch difficult problems. Evaluation on pedestrian tracking is provided for\nmultiple scenarios, showing superior results over single detector tracking and\nstandard QP-solvers. Finally, our tracker ranks 2nd on the MOT16 benchmark and\n1st on the new MOT17 benchmark, outperforming over 90 trackers.","url_abs":"http://arxiv.org/abs/1705.08314v4","url_pdf":"http://arxiv.org/pdf/1705.08314v4.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":[],"tasks":[{"task_slug":"multi-object-tracking","task_name":"Multi-Object Tracking"},{"task_slug":"object-tracking","task_name":"Object Tracking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/multi-object-tracking-on-mot16","task":"Multi-Object Tracking","dataset":"MOT16","model":"FWT","rank_in_archive_order":22,"of":24,"metrics":{"MOTA":"47.8"},"uses_additional_data":false},{"leaderboard":"/sota/multi-object-tracking-on-mot17","task":"Multi-Object Tracking","dataset":"MOT17","model":"FWT","rank_in_archive_order":47,"of":48,"metrics":{"MOTA":"51.3"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1705.08314","atlas_url":"https://app.syntology.ai/?focus=1705.08314","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}