9 citations · 9 across the 5 of their papers we have counts for
1 paper · 1 filter
Yanjie Li, Mingxing Duan, Xuelong Dai +1
Adversarial patch attacks can fool the face recognition (FR) models via small patches. However, previous adversarial patch attacks often result in unnatural patterns that are easil…