文献类型：期刊文献

英文题名：A Face Spoofing Detection Method Based on Domain Adaptation and Lossless Size Adaptation

作者：Sun, Wenyun[1,2,3];Song, Yu[1,2,3];Zhao, Haitao[4];Jin, Zhong[5]

机构：[1]Shenzhen Univ, Coll Elect & Informat Engn, Shenzhen 518060, Peoples R China;[2]Shenzhen Univ, Shenzhen Key Lab Media Secur, Shenzhen 518060, Peoples R China;[3]Shenzhen Univ, Guangdong Key Lab Intelligent Informat Proc, Shenzhen 518060, Peoples R China;[4]East China Univ Sci & Technol, Sch Informat Sci & Engn, Shanghai 200237, Peoples R China;[5]Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Nanjing 210094, Peoples R China

年份：2020

卷号：8

起止页码：66553

外文期刊名：IEEE ACCESS

收录：;EI(收录号：20201808589091);WOS:【SCI-EXPANDED(收录号:WOS:000527415800015)】；

基金：This work was supported in part by the National Natural Science Foundation of China under Grant 61902250, and in part by the China Postdoctoral Science Foundation under Grant 2018M643183.

语种：英文

外文关键词：Domain adaptation; face anti-spoofing; face liveness detection; face presentation attack detection; face spoofing detection; forensics; machine learning; pattern recognition

摘要：In this paper, a face spoofing detection method called the Fully Convolutional Network with Domain Adaptation and Lossless Size Adaptation (FCN-DA-LSA) is proposed. As its name suggests, the FCN-DA-LSA includes a lossless size adaptation preprocessor followed by an FCN based pixel-level classifier embedded with a domain adaptation layer. The FCN local classifier makes full use of the basic properties of face spoof distortion namely ubiquitous and repetitive. The domain adaptation (DA) layer improves generalization across different domains. The lossless size adaptation (LSA) preserves the high-frequent spoof clues caused by the face recapture process. The ablation study shows that both DA and the LSA are necessary for high-accuracy face spoofing detection. The FCN-LSA obtains competitive performance among the state-of-the-art methods. With the help of small-sample external data in the target domain (2/50, 2/50, and 1/20 subjects for CASIA-FASD, Replay-Attack, and OULU-NPU respectively), the FCN-DA-LSA further improves the performance and outperforms the existing methods.