Universal Adversarial Spoofing Attacks against Face Recognition

Autor:	Amada, Takuma, Liew, Seng Pei, Kakizaki, Kazuya, Araki, Toshinori
Rok vydání:	2021
Předmět:	Computer Science - Computer Vision and Pattern Recognition Computer Science - Cryptography and Security Computer Science - Machine Learning
Druh dokumentu:	Working Paper
DOI:	10.1109/IJCB52358.2021.9484380
Popis:	We assess the vulnerabilities of deep face recognition systems for images that falsify/spoof multiple identities simultaneously. We demonstrate that, by manipulating the deep feature representation extracted from a face image via imperceptibly small perturbations added at the pixel level using our proposed Universal Adversarial Spoofing Examples (UAXs), one can fool a face verification system into recognizing that the face image belongs to multiple different identities with a high success rate. One characteristic of the UAXs crafted with our method is that they are universal (identity-agnostic); they are successful even against identities not known in advance. For a certain deep neural network, we show that we are able to spoof almost all tested identities (99\%), including those not known beforehand (not included in training). Our results indicate that a multiple-identity attack is a real threat and should be taken into account when deploying face recognition systems. Comment: Accepted to International Joint Conference on Biometrics (IJCB 2021)
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2110.00708 Zobrazit plný text záznamu View this record from Arxiv