{"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/visual-object-tracking-the-initialisation","title":"Visual Object Tracking: The Initialisation Problem","arxiv_id":"1805.01146","date":"2018-05-03","proceeding":null,"authors":["George De Ath","Richard Everson"],"abstract":"Model initialisation is an important component of object tracking. Tracking\nalgorithms are generally provided with the first frame of a sequence and a\nbounding box (BB) indicating the location of the object. This BB may contain a\nlarge number of background pixels in addition to the object and can lead to\nparts-based tracking algorithms initialising their object models in background\nregions of the BB. In this paper, we tackle this as a missing labels problem,\nmarking pixels sufficiently away from the BB as belonging to the background and\nlearning the labels of the unknown pixels. Three techniques, One-Class SVM\n(OC-SVM), Sampled-Based Background Model (SBBM) (a novel background model based\non pixel samples), and Learning Based Digital Matting (LBDM), are adapted to\nthe problem. These are evaluated with leave-one-video-out cross-validation on\nthe VOT2016 tracking benchmark. Our evaluation shows both OC-SVMs and SBBM are\ncapable of providing a good level of segmentation accuracy but are too\nparameter-dependent to be used in real-world scenarios. We show that LBDM\nachieves significantly increased performance with parameters selected by cross\nvalidation and we show that it is robust to parameter variation.","url_abs":"http://arxiv.org/abs/1805.01146v2","url_pdf":"http://arxiv.org/pdf/1805.01146v2.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":"visual-object-tracking-the-initialisation","repo_url":"https://github.com/georgedeath/initialisation-problem","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"image-matting","task_name":"Image Matting"},{"task_slug":"missing-labels","task_name":"Missing Labels"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"visual-object-tracking","task_name":"Visual Object Tracking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}