-
-
Identifying real image
-
UI
-
Detecting fake
Inspiration
To combat the growing misuse of deepfakes targeting women by offering a fast, reliable system to verify image authenticity and build digital trust.
What it does
Classifies images as REAL or AI-generated using a lightweight ML model, with a simple web interface.
How we built it
Python backend TensorFlow + MobileNetV2 for model MTCNN for face detection Flask for web interface Pillow & Matplotlib for image handling
Challenges we ran into
Handling small datasets and preprocessing images TensorFlow environment setup on Windows
Accomplishments that we're proud of
Working AI-powered deepfake detector Web interface for real-time predictions
What we learned
Preprocessing pipelines and face detection Transfer learning with MobileNetV2 Flask deployment and handling AI models in web app
What's next for SafeFace
Improve model accuracy with larger datasets Add video-based deepfake detection Deploy online for public use
Log in or sign up for Devpost to join the conversation.