{"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/beyond-pixels-leveraging-geometry-and-shape","title":"Beyond Pixels: Leveraging Geometry and Shape Cues for Online Multi-Object Tracking","arxiv_id":"1802.09298","date":"2018-02-26","proceeding":null,"authors":["Sarthak Sharma","Junaid Ahmed Ansari","J. Krishna Murthy","K. Madhava Krishna"],"abstract":"This paper introduces geometry and object shape and pose costs for\nmulti-object tracking in urban driving scenarios. Using images from a monocular\ncamera alone, we devise pairwise costs for object tracks, based on several 3D\ncues such as object pose, shape, and motion. The proposed costs are agnostic to\nthe data association method and can be incorporated into any optimization\nframework to output the pairwise data associations. These costs are easy to\nimplement, can be computed in real-time, and complement each other to account\nfor possible errors in a tracking-by-detection framework. We perform an\nextensive analysis of the designed costs and empirically demonstrate consistent\nimprovement over the state-of-the-art under varying conditions that employ a\nrange of object detectors, exhibit a variety in camera and object motions, and,\nmore importantly, are not reliant on the choice of the association framework.\nWe also show that, by using the simplest of associations frameworks (two-frame\nHungarian assignment), we surpass the state-of-the-art in multi-object-tracking\non road scenes. More qualitative and quantitative results can be found at the\nfollowing URL: https://junaidcs032.github.io/Geometry_ObjectShape_MOT/.","url_abs":"http://arxiv.org/abs/1802.09298v2","url_pdf":"http://arxiv.org/pdf/1802.09298v2.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":"beyond-pixels-leveraging-geometry-and-shape","repo_url":"https://github.com/JunaidCS032/MOTBeyondPixels","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"multi-object-tracking","task_name":"Multi-Object Tracking"},{"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":"online-multi-object-tracking","task_name":"Online Multi-Object Tracking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/multiple-object-tracking-on-kitti-test-online","task":"Multiple Object Tracking","dataset":"KITTI Test (Online Methods)","model":"RRC-IIITH","rank_in_archive_order":25,"of":34,"metrics":{"MOTA":"84.24"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.09298","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}