{"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/long-term-visual-object-tracking-benchmark","title":"Long-Term Visual Object Tracking Benchmark","arxiv_id":"1712.01358","date":"2017-12-04","proceeding":null,"authors":["Abhinav Moudgil","Vineet Gandhi"],"abstract":"We propose a new long video dataset (called Track Long and Prosper - TLP) and\nbenchmark for single object tracking. The dataset consists of 50 HD videos from\nreal world scenarios, encompassing a duration of over 400 minutes (676K\nframes), making it more than 20 folds larger in average duration per sequence\nand more than 8 folds larger in terms of total covered duration, as compared to\nexisting generic datasets for visual tracking. The proposed dataset paves a way\nto suitably assess long term tracking performance and train better deep\nlearning architectures (avoiding/reducing augmentation, which may not reflect\nreal world behaviour). We benchmark the dataset on 17 state of the art trackers\nand rank them according to tracking accuracy and run time speeds. We further\npresent thorough qualitative and quantitative evaluation highlighting the\nimportance of long term aspect of tracking. Our most interesting observations\nare (a) existing short sequence benchmarks fail to bring out the inherent\ndifferences in tracking algorithms which widen up while tracking on long\nsequences and (b) the accuracy of trackers abruptly drops on challenging long\nsequences, suggesting the potential need of research efforts in the direction\nof long-term tracking.","url_abs":"http://arxiv.org/abs/1712.01358v4","url_pdf":"http://arxiv.org/pdf/1712.01358v4.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":"long-term-visual-object-tracking-benchmark","repo_url":"https://github.com/JoseVillagranE/SiameseNetworks","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"visual-object-tracking","task_name":"Visual Object Tracking"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"}],"methods":[],"datasets_introduced":[{"slug":"tlp","name":"TLP","full_name":"Track Long and Prosper"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.01358","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}