{"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/mobiface-a-novel-dataset-for-mobile-face","title":"MobiFace: A Novel Dataset for Mobile Face Tracking in the Wild","arxiv_id":"1805.09749","date":"2018-05-24","proceeding":null,"authors":["Yiming Lin","Shiyang Cheng","Jie Shen","Maja Pantic"],"abstract":"Face tracking serves as the crucial initial step in mobile applications\ntrying to analyse target faces over time in mobile settings. However, this\nproblem has received little attention, mainly due to the scarcity of dedicated\nface tracking benchmarks. In this work, we introduce MobiFace, the first\ndataset for single face tracking in mobile situations. It consists of 80\nunedited live-streaming mobile videos captured by 70 different smartphone users\nin fully unconstrained environments. Over $95K$ bounding boxes are manually\nlabelled. The videos are carefully selected to cover typical smartphone usage.\nThe videos are also annotated with 14 attributes, including 6 newly proposed\nattributes and 8 commonly seen in object tracking. 36 state-of-the-art\ntrackers, including facial landmark trackers, generic object trackers and\ntrackers that we have fine-tuned or improved, are evaluated. The results\nsuggest that mobile face tracking cannot be solved through existing approaches.\nIn addition, we show that fine-tuning on the MobiFace training data\nsignificantly boosts the performance of deep learning-based trackers,\nsuggesting that MobiFace captures the unique characteristics of mobile face\ntracking. Our goal is to offer the community a diverse dataset to enable the\ndesign and evaluation of mobile face trackers. The dataset, annotations and the\nevaluation server will be on \\url{https://mobiface.github.io/}.","url_abs":"http://arxiv.org/abs/1805.09749v2","url_pdf":"http://arxiv.org/pdf/1805.09749v2.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":"mobiface-a-novel-dataset-for-mobile-face","repo_url":"https://github.com/mobiface/mobiface.github.io","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"face-detection","task_name":"Face Detection"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"}],"methods":[],"datasets_introduced":[{"slug":"mobiface","name":"MobiFace","full_name":"MobiFace"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}