Spoofing attacks are one of the biggest concerns for most biometric recognition systems. This will be also the case with silhouette-based gait recognition in the near future. So far, gait recognition has been fortunately out of the scope of spoofing attacks. However, it is becoming a real threat with the rapid growth and spread of deep neural network-based multimedia generation techniques, which will allow attackers to generate a fake video of gait silhouettes resembling a target person's walking motion. We refer to such computer-generated fake silhouettes as gait silhouette clones (GSCs). To deal with the future threat caused by GSCs, in this paper, we propose a supervised method for discriminating GSCs from genuine gait silhouettes (GGSs) that are observed from actual walking people. For training a good discriminator, it is important to collect training datasets of both GGSs and GSCs which do not differ from each other in any aspect other than genuineness. To this end, we propose to generate a training set of GSCs from GGSs by transforming them using multiple autoencoders. The generated GSCs are used together with their original GGSs for training the discriminator. In our experiments, the proposed method achieved the recognition accuracy of up to 94% for several test datasets, which demonstrates the effectiveness and the generality of the proposed method.
Yuki HIROSE
Osaka University
Kazuaki NAKAMURA
Osaka University
Naoko NITTA
Osaka University
Noboru BABAGUCHI
Osaka University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yuki HIROSE, Kazuaki NAKAMURA, Naoko NITTA, Noboru BABAGUCHI, "Discrimination between Genuine and Cloned Gait Silhouette Videos via Autoencoder-Based Training Data Generation" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 12, pp. 2535-2546, December 2019, doi: 10.1587/transinf.2019EDP7042.
Abstract: Spoofing attacks are one of the biggest concerns for most biometric recognition systems. This will be also the case with silhouette-based gait recognition in the near future. So far, gait recognition has been fortunately out of the scope of spoofing attacks. However, it is becoming a real threat with the rapid growth and spread of deep neural network-based multimedia generation techniques, which will allow attackers to generate a fake video of gait silhouettes resembling a target person's walking motion. We refer to such computer-generated fake silhouettes as gait silhouette clones (GSCs). To deal with the future threat caused by GSCs, in this paper, we propose a supervised method for discriminating GSCs from genuine gait silhouettes (GGSs) that are observed from actual walking people. For training a good discriminator, it is important to collect training datasets of both GGSs and GSCs which do not differ from each other in any aspect other than genuineness. To this end, we propose to generate a training set of GSCs from GGSs by transforming them using multiple autoencoders. The generated GSCs are used together with their original GGSs for training the discriminator. In our experiments, the proposed method achieved the recognition accuracy of up to 94% for several test datasets, which demonstrates the effectiveness and the generality of the proposed method.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019EDP7042/_p
Copy
@ARTICLE{e102-d_12_2535,
author={Yuki HIROSE, Kazuaki NAKAMURA, Naoko NITTA, Noboru BABAGUCHI, },
journal={IEICE TRANSACTIONS on Information},
title={Discrimination between Genuine and Cloned Gait Silhouette Videos via Autoencoder-Based Training Data Generation},
year={2019},
volume={E102-D},
number={12},
pages={2535-2546},
abstract={Spoofing attacks are one of the biggest concerns for most biometric recognition systems. This will be also the case with silhouette-based gait recognition in the near future. So far, gait recognition has been fortunately out of the scope of spoofing attacks. However, it is becoming a real threat with the rapid growth and spread of deep neural network-based multimedia generation techniques, which will allow attackers to generate a fake video of gait silhouettes resembling a target person's walking motion. We refer to such computer-generated fake silhouettes as gait silhouette clones (GSCs). To deal with the future threat caused by GSCs, in this paper, we propose a supervised method for discriminating GSCs from genuine gait silhouettes (GGSs) that are observed from actual walking people. For training a good discriminator, it is important to collect training datasets of both GGSs and GSCs which do not differ from each other in any aspect other than genuineness. To this end, we propose to generate a training set of GSCs from GGSs by transforming them using multiple autoencoders. The generated GSCs are used together with their original GGSs for training the discriminator. In our experiments, the proposed method achieved the recognition accuracy of up to 94% for several test datasets, which demonstrates the effectiveness and the generality of the proposed method.},
keywords={},
doi={10.1587/transinf.2019EDP7042},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - Discrimination between Genuine and Cloned Gait Silhouette Videos via Autoencoder-Based Training Data Generation
T2 - IEICE TRANSACTIONS on Information
SP - 2535
EP - 2546
AU - Yuki HIROSE
AU - Kazuaki NAKAMURA
AU - Naoko NITTA
AU - Noboru BABAGUCHI
PY - 2019
DO - 10.1587/transinf.2019EDP7042
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2019
AB - Spoofing attacks are one of the biggest concerns for most biometric recognition systems. This will be also the case with silhouette-based gait recognition in the near future. So far, gait recognition has been fortunately out of the scope of spoofing attacks. However, it is becoming a real threat with the rapid growth and spread of deep neural network-based multimedia generation techniques, which will allow attackers to generate a fake video of gait silhouettes resembling a target person's walking motion. We refer to such computer-generated fake silhouettes as gait silhouette clones (GSCs). To deal with the future threat caused by GSCs, in this paper, we propose a supervised method for discriminating GSCs from genuine gait silhouettes (GGSs) that are observed from actual walking people. For training a good discriminator, it is important to collect training datasets of both GGSs and GSCs which do not differ from each other in any aspect other than genuineness. To this end, we propose to generate a training set of GSCs from GGSs by transforming them using multiple autoencoders. The generated GSCs are used together with their original GGSs for training the discriminator. In our experiments, the proposed method achieved the recognition accuracy of up to 94% for several test datasets, which demonstrates the effectiveness and the generality of the proposed method.
ER -