Adversarial Lab

Adversarial Lab

This project is a Web-based Tool for visualisation and generation of adversarial examples by attacking ImageNet Models like VGG, AlexNet, ResNet etc.

Visualizing and Comparision of Various Adversarial Attacks on user uploaded images using a simple interface, using the DNN framework Pytorch, using popular SOTA Pretrained TorchVision ModelZoo. The Following Attacks have been implemented so far:

FGSM

Fast Gradient Sign Method, Untargeted Fast Gradient Sign Method, Targeted Iterative

Basic Iterative Method, Untargeted Least Likely Class Iterative Method DeepFool, untargeted

LBFGS, targeted

Coming Soon: Carlini-Wagner l2, and Many More