A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios
Face Anti-spoofing (FAS) is a challenging problem due to complex serving scenarios and diverse face presentation attack patterns. Especially when captured images are low-resolution, blurry, and coming from different domains, the performance of FAS will degrade significantly. The existing multi-modal...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2021-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9328436/ |
id |
doaj-35784957152c4b5db1c2c29468b477b6 |
---|---|
record_format |
Article |
spelling |
doaj-35784957152c4b5db1c2c29468b477b62021-03-30T15:26:17ZengIEEEIEEE Access2169-35362021-01-019281402815510.1109/ACCESS.2021.30527289328436A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance ScenariosXudong Chen0https://orcid.org/0000-0002-6791-6182Shugong Xu1https://orcid.org/0000-0003-1905-6269Qiaobin Ji2Shan Cao3https://orcid.org/0000-0003-3713-8671Shanghai Institute for Advanced Communication and Data Science, Shanghai University, Shanghai, ChinaShanghai Institute for Advanced Communication and Data Science, Shanghai University, Shanghai, ChinaShanghai Institute for Advanced Communication and Data Science, Shanghai University, Shanghai, ChinaShanghai Institute for Advanced Communication and Data Science, Shanghai University, Shanghai, ChinaFace Anti-spoofing (FAS) is a challenging problem due to complex serving scenarios and diverse face presentation attack patterns. Especially when captured images are low-resolution, blurry, and coming from different domains, the performance of FAS will degrade significantly. The existing multi-modal FAS datasets rarely pay attention to the cross-domain problems under deployment scenarios, which is not conducive to the study of model performance. To solve these problems, we explore the fine-grained differences between multi-modal cameras and construct a cross-domain multi-modal FAS dataset under surveillance scenarios called GREAT-FASD-S. Besides, we propose an Attention based Face Anti-spoofing network with Feature Augment (AFA) to solve the FAS towards low-quality face images. It consists of the depthwise separable attention module (DAM) and the multi-modal based feature augment module (MFAM). Our model can achieve state-of-the-art performance on the CASIA-SURF dataset and our proposed GREAT-FASD-S dataset.https://ieeexplore.ieee.org/document/9328436/Face anti-spoofingmulti-modalsurveillance scenarioscross domain |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Xudong Chen Shugong Xu Qiaobin Ji Shan Cao |
spellingShingle |
Xudong Chen Shugong Xu Qiaobin Ji Shan Cao A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios IEEE Access Face anti-spoofing multi-modal surveillance scenarios cross domain |
author_facet |
Xudong Chen Shugong Xu Qiaobin Ji Shan Cao |
author_sort |
Xudong Chen |
title |
A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios |
title_short |
A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios |
title_full |
A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios |
title_fullStr |
A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios |
title_full_unstemmed |
A Dataset and Benchmark Towards Multi-Modal Face Anti-Spoofing Under Surveillance Scenarios |
title_sort |
dataset and benchmark towards multi-modal face anti-spoofing under surveillance scenarios |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2021-01-01 |
description |
Face Anti-spoofing (FAS) is a challenging problem due to complex serving scenarios and diverse face presentation attack patterns. Especially when captured images are low-resolution, blurry, and coming from different domains, the performance of FAS will degrade significantly. The existing multi-modal FAS datasets rarely pay attention to the cross-domain problems under deployment scenarios, which is not conducive to the study of model performance. To solve these problems, we explore the fine-grained differences between multi-modal cameras and construct a cross-domain multi-modal FAS dataset under surveillance scenarios called GREAT-FASD-S. Besides, we propose an Attention based Face Anti-spoofing network with Feature Augment (AFA) to solve the FAS towards low-quality face images. It consists of the depthwise separable attention module (DAM) and the multi-modal based feature augment module (MFAM). Our model can achieve state-of-the-art performance on the CASIA-SURF dataset and our proposed GREAT-FASD-S dataset. |
topic |
Face anti-spoofing multi-modal surveillance scenarios cross domain |
url |
https://ieeexplore.ieee.org/document/9328436/ |
work_keys_str_mv |
AT xudongchen adatasetandbenchmarktowardsmultimodalfaceantispoofingundersurveillancescenarios AT shugongxu adatasetandbenchmarktowardsmultimodalfaceantispoofingundersurveillancescenarios AT qiaobinji adatasetandbenchmarktowardsmultimodalfaceantispoofingundersurveillancescenarios AT shancao adatasetandbenchmarktowardsmultimodalfaceantispoofingundersurveillancescenarios AT xudongchen datasetandbenchmarktowardsmultimodalfaceantispoofingundersurveillancescenarios AT shugongxu datasetandbenchmarktowardsmultimodalfaceantispoofingundersurveillancescenarios AT qiaobinji datasetandbenchmarktowardsmultimodalfaceantispoofingundersurveillancescenarios AT shancao datasetandbenchmarktowardsmultimodalfaceantispoofingundersurveillancescenarios |
_version_ |
1724179388810919936 |