Fig. 2From: HVS-inspired adversarial image generation with high perceptual qualityComparison among five anchor methods and the proposed method under targeted and non-targeted attack on face recognition. Anchor methods include BIM [21], DI\(^2\)FGSM [20], MIFGSM [26], MND [29], JNDMGA [28] and PGD [25]. A specific part in each image is enlarged in the red boxBack to article page