{"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/casia-surf-a-dataset-and-benchmark-for-large","title":"A Dataset and Benchmark for Large-scale Multi-modal Face Anti-spoofing","arxiv_id":"1812.00408","date":"2018-12-02","proceeding":"CVPR 2019 6","authors":["Shifeng Zhang","Xiaobo Wang","Ajian Liu","Chenxu Zhao","Jun Wan","Sergio Escalera","Hailin Shi","Zezheng Wang","Stan Z. Li"],"abstract":"Face anti-spoofing is essential to prevent face recognition systems from a\nsecurity breach. Much of the progresses have been made by the availability of\nface anti-spoofing benchmark datasets in recent years. However, existing face\nanti-spoofing benchmarks have limited number of subjects ($\\le\\negmedspace170$)\nand modalities ($\\leq\\negmedspace2$), which hinder the further development of\nthe academic community. To facilitate face anti-spoofing research, we introduce\na large-scale multi-modal dataset, namely CASIA-SURF, which is the largest\npublicly available dataset for face anti-spoofing in terms of both subjects and\nvisual modalities. Specifically, it consists of $1,000$ subjects with $21,000$\nvideos and each sample has $3$ modalities (i.e., RGB, Depth and IR). We also\nprovide a measurement set, evaluation protocol and training/validation/testing\nsubsets, developing a new benchmark for face anti-spoofing. Moreover, we\npresent a new multi-modal fusion method as baseline, which performs feature\nre-weighting to select the more informative channel features while suppressing\nthe less useful ones for each modal. Extensive experiments have been conducted\non the proposed dataset to verify its significance and generalization\ncapability. The dataset is available at\nhttps://sites.google.com/qq.com/chalearnfacespoofingattackdete","url_abs":"http://arxiv.org/abs/1812.00408v3","url_pdf":"http://arxiv.org/pdf/1812.00408v3.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":"casia-surf-a-dataset-and-benchmark-for-large","repo_url":"https://github.com/SoftwareGift/FeatherNets_Face-Anti-spoofing-Attack-Detection-Challenge-CVPR2019","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"casia-surf-a-dataset-and-benchmark-for-large","repo_url":"https://github.com/avuku06/G","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"face-anti-spoofing","task_name":"Face Anti-Spoofing"},{"task_slug":"face-recognition","task_name":"Face Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1812.00408","atlas_url":"https://app.syntology.ai/?focus=1812.00408","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}