{"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/multi-camera-action-dataset-for-cross-camera","title":"Multi-Camera Action Dataset for Cross-Camera Action Recognition Benchmarking","arxiv_id":"1607.06408","date":"2016-07-21","proceeding":null,"authors":["Wenhui Li","Yongkang Wong","An-An Liu","Yang Li","Yu-Ting Su","Mohan Kankanhalli"],"abstract":"Action recognition has received increasing attention from the computer vision\nand machine learning communities in the last decade. To enable the study of\nthis problem, there exist a vast number of action datasets, which are recorded\nunder controlled laboratory settings, real-world surveillance environments, or\ncrawled from the Internet. Apart from the \"in-the-wild\" datasets, the training\nand test split of conventional datasets often possess similar environments\nconditions, which leads to close to perfect performance on constrained\ndatasets. In this paper, we introduce a new dataset, namely Multi-Camera Action\nDataset (MCAD), which is designed to evaluate the open view classification\nproblem under the surveillance environment. In total, MCAD contains 14,298\naction samples from 18 action categories, which are performed by 20 subjects\nand independently recorded with 5 cameras. Inspired by the well received\nevaluation approach on the LFW dataset, we designed a standard evaluation\nprotocol and benchmarked MCAD under several scenarios. The benchmark shows that\nwhile an average of 85% accuracy is achieved under the closed-view scenario,\nthe performance suffers from a significant drop under the cross-view scenario.\nIn the worst case scenario, the performance of 10-fold cross validation drops\nfrom 87.0% to 47.4%.","url_abs":"http://arxiv.org/abs/1607.06408v3","url_pdf":"http://arxiv.org/pdf/1607.06408v3.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":[],"tasks":[{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"}],"methods":[],"datasets_introduced":[{"slug":"mcad","name":"MCAD","full_name":"Multi-Camera Action Dataset"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}