Skip to content

Latest commit

 

History

History
22 lines (16 loc) · 771 Bytes

README.md

File metadata and controls

22 lines (16 loc) · 771 Bytes

SAPA

Repository of Scale-Agnostic Photorealistic Adversarial Pattern Attack

Attack

  • Please download the weights and put them under SAPA/models/pretrained.

Classification model

  • Put the images to SAPA/test_imgs and specify their ground truth label under SAPA/test_imgs/labels.txt.

  • Run python attack.py

  • Attack results will be saved to SAPA/attack_results.

    (Notice that the images can be in any shape but please make sure the configurations in attack.py are changed correspondingly.)

Object detection model

Coming soon.

PQGAN Training

Coming Soon.