TY - GEN
T1 - MAP
T2 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2022
AU - Kim, Taeheon
AU - Lee, Hong Joo
AU - Ro, Yong Man
N1 - Publisher Copyright:
© 2022 IEEE
PY - 2022
Y1 - 2022
N2 - Recently, multispectral person detection has shown great performance in real world applications such as autonomous driving and security systems. However, the reliability of person detection against physical attacks has not been fully explored yet in multispectral person detectors. To evaluate the robustness of multispectral person detectors in the physical world, we propose a novel Multispectral Adversarial Patch (MAP) generation framework. MAP is optimized with a Cross-spectral Mapping(CSM) and Material Emissivity(ME) loss. This paper is the first to evaluate the reliability of a multispectral person detector against physical attack. Throughout experiment, our proposed adversarial patch successfully attacks the person detector and the Average Precision (AP) score is dropped by 90.79% in digital space and 73.34% in physical space.
AB - Recently, multispectral person detection has shown great performance in real world applications such as autonomous driving and security systems. However, the reliability of person detection against physical attacks has not been fully explored yet in multispectral person detectors. To evaluate the robustness of multispectral person detectors in the physical world, we propose a novel Multispectral Adversarial Patch (MAP) generation framework. MAP is optimized with a Cross-spectral Mapping(CSM) and Material Emissivity(ME) loss. This paper is the first to evaluate the reliability of a multispectral person detector against physical attack. Throughout experiment, our proposed adversarial patch successfully attacks the person detector and the Average Precision (AP) score is dropped by 90.79% in digital space and 73.34% in physical space.
KW - Cross-spectral mapping
KW - Material emissivity loss
KW - Multispectral adversarial patch
KW - Person detection
KW - Thermal adversarial patch
UR - https://www.scopus.com/pages/publications/85134021630
U2 - 10.1109/ICASSP43922.2022.9747896
DO - 10.1109/ICASSP43922.2022.9747896
M3 - Conference contribution
AN - SCOPUS:85134021630
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 4853
EP - 4857
BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 22 May 2022 through 27 May 2022
ER -