{"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/pathtrack-fast-trajectory-annotation-with","title":"PathTrack: Fast Trajectory Annotation with Path Supervision","arxiv_id":"1703.02437","date":"2017-03-07","proceeding":"ICCV 2017 10","authors":["Santiago Manen","Michael Gygli","Dengxin Dai","Luc van Gool"],"abstract":"Progress in Multiple Object Tracking (MOT) has been historically limited by\nthe size of the available datasets. We present an efficient framework to\nannotate trajectories and use it to produce a MOT dataset of unprecedented\nsize. In our novel path supervision the annotator loosely follows the object\nwith the cursor while watching the video, providing a path annotation for each\nobject in the sequence. Our approach is able to turn such weak annotations into\ndense box trajectories. Our experiments on existing datasets prove that our\nframework produces more accurate annotations than the state of the art, in a\nfraction of the time. We further validate our approach by crowdsourcing the\nPathTrack dataset, with more than 15,000 person trajectories in 720 sequences.\nTracking approaches can benefit training on such large-scale datasets, as did\nobject recognition. We prove this by re-training an off-the-shelf person\nmatching network, originally trained on the MOT15 dataset, almost halving the\nmisclassification rate. Additionally, training on our data consistently\nimproves tracking results, both on our dataset and on MOT15. On the latter, we\nimprove the top-performing tracker (NOMT) dropping the number of IDSwitches by\n18% and fragments by 5%.","url_abs":"http://arxiv.org/abs/1703.02437v2","url_pdf":"http://arxiv.org/pdf/1703.02437v2.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":"multiple-object-tracking","task_name":"Multiple Object Tracking"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-recognition","task_name":"Object Recognition"},{"task_slug":"object-tracking","task_name":"Object Tracking"}],"methods":[],"datasets_introduced":[{"slug":"pathtrack","name":"PathTrack","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.02437","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}