Targeted Adversarial Attack using FGSM Method in Facial Recognition Embedding Model

Usage Guide • Contributors • Quick Start

FaceGSM designed for performing targeted adversarial attacks using the FGSM (Fast Gradient Sign Method) in Facial Recognition Embedding Model. FaceGSM revolutionizes security testing with a suite of innovative features, including:

• Static - Takes static images as input for FaceGSM.
• Capture - Takes image captured by camera as input for FaceGSM
• Live - Takes real-time live video feed frames as input for FaceGSM.

Attacker's Face (Clario)	Target's Face (Clints)	Output : Generated Adversarial Image	Attack Result : Attacker's Face Predicted as Victim

[🔑] Key Features

✅ Fully compatible with multiple facial recognition embedding model including FaceNet, ArcFace, GhostFaceNet, DeepID, and VGGFace2
✅ Supports multiple input media, including static image, captured image and live video feed
✅ Saved generated adversarial image as checkpoints to increase efficiency for future attacks
✅ Works with your own Custom Face Datasets
✅ Provide easy installation and intuitive UI/UX

[⚙️] Installation

Clone the Repository

# Clone the FaceGSM repository
$ git clone https://github.com/facegsmproject/FaceGSM

Conda Environment

Install Conda from here. After installing Conda, Run the following commands :

# Create the environment for FaceGSM using the provided facegsm.yml file
$ conda env create --name facegsm --file facegsm.yml

# Activate the FaceGSM environment
$ conda activate facegsm

For more detail about the installation, please refer to our Installation Guide

[⌛] Quick Start

To test if FaceGSM is working properly you can use FaceGSM's default datasets and run the following command:

$ python3 facegsm.py static --original ./datasets/04.jpg --target ./datasets/74.jpg

Help

$ python3 facegsm.py --help

Usage: python3 facegsm.py [ static | capture | live | database ] --help
Options:
  static: Static input for FGSM attack in FaceGSM.
  capture: Capture original and target photos in FaceGSM.
  live: Live camera feature in FaceGSM includes real-time face recognition and attack capabilities.
  database: Create a database based on datasets for FaceGSM.
  --help: Show help for available options.

Custom Face Datasets

Generate your own custom face datasets by running the following command:

$ python3 facegsm.py database --dataset ./your_custom_dataset_folder

Run FaceGSM with your custom face datasets:

$ python3 facegsm.py static --original ./your_custom_dataset_folder/a.jpg --target ./your_custom_dataset_folder/b.jpg

Credits

FaceGSM is developed by :

• Excy
• Fejka
• Maskirovka