{"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/need-for-speed-a-benchmark-for-higher-frame","title":"Need for Speed: A Benchmark for Higher Frame Rate Object Tracking","arxiv_id":"1703.05884","date":"2017-03-17","proceeding":"ICCV 2017 10","authors":["Hamed Kiani Galoogahi","Ashton Fagg","Chen Huang","Deva Ramanan","Simon Lucey"],"abstract":"In this paper, we propose the first higher frame rate video dataset (called\nNeed for Speed - NfS) and benchmark for visual object tracking. The dataset\nconsists of 100 videos (380K frames) captured with now commonly available\nhigher frame rate (240 FPS) cameras from real world scenarios. All frames are\nannotated with axis aligned bounding boxes and all sequences are manually\nlabelled with nine visual attributes - such as occlusion, fast motion,\nbackground clutter, etc. Our benchmark provides an extensive evaluation of many\nrecent and state-of-the-art trackers on higher frame rate sequences. We ranked\neach of these trackers according to their tracking accuracy and real-time\nperformance. One of our surprising conclusions is that at higher frame rates,\nsimple trackers such as correlation filters outperform complex methods based on\ndeep networks. This suggests that for practical applications (such as in\nrobotics or embedded vision), one needs to carefully tradeoff bandwidth\nconstraints associated with higher frame rate acquisition, computational costs\nof real-time analysis, and the required application accuracy. Our dataset and\nbenchmark allows for the first time (to our knowledge) systematic exploration\nof such issues, and will be made available to allow for further research in\nthis space.","url_abs":"http://arxiv.org/abs/1703.05884v2","url_pdf":"http://arxiv.org/pdf/1703.05884v2.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":"need-for-speed-a-benchmark-for-higher-frame","repo_url":"https://github.com/susomena/DeepSlowMotion","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"visual-object-tracking","task_name":"Visual Object Tracking"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1703.05884","atlas_url":"https://app.syntology.ai/?focus=1703.05884","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}