{"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/detection-tracking-for-efficient-person","title":"Detection-Tracking for Efficient Person Analysis: The DetTA Pipeline","arxiv_id":"1804.10134","date":"2018-04-26","proceeding":null,"authors":["Stefan Breuers","Lucas Beyer","Umer Rafi","Bastian Leibe"],"abstract":"In the past decade many robots were deployed in the wild, and people\ndetection and tracking is an important component of such deployments. On top of\nthat, one often needs to run modules which analyze persons and extract higher\nlevel attributes such as age and gender, or dynamic information like gaze and\npose. The latter ones are especially necessary for building a reactive, social\nrobot-person interaction.\n  In this paper, we combine those components in a fully modular\ndetection-tracking-analysis pipeline, called DetTA. We investigate the benefits\nof such an integration on the example of head and skeleton pose, by using the\nconsistent track ID for a temporal filtering of the analysis modules'\nobservations, showing a slight improvement in a challenging real-world\nscenario. We also study the potential of a so-called \"free-flight\" mode, where\nthe analysis of a person attribute only relies on the filter's predictions for\ncertain frames. Here, our study shows that this boosts the runtime\ndramatically, while the prediction quality remains stable. This insight is\nespecially important for reducing power consumption and sharing precious\n(GPU-)memory when running many analysis components on a mobile platform,\nespecially so in the era of expensive deep learning methods.","url_abs":"http://arxiv.org/abs/1804.10134v2","url_pdf":"http://arxiv.org/pdf/1804.10134v2.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":"detection-tracking-for-efficient-person","repo_url":"https://github.com/sbreuers/detta","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":null,"task_name":"GPU"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}