Deepfake Detection Research
Deepfake Detection Research
A face-swap detection model, trained on an A40 GPU and tested on a Raspberry Pi.
A deep learning model that detects face-swap deepfakes, built during a research internship at Universiti Sains Malaysia. It was trained on an NVIDIA A40 GPU, with inference tested on a Raspberry Pi.
Features
Datasets
Curated and prepared the face-swap datasets for training.
Training
Experiments run on an NVIDIA A40 GPU server.
Edge inference
Inference validated on a Raspberry Pi.
Research paper
Findings co-authored in an academic paper.
Tools
- Python
- Deep Learning
- Computer Vision
- Edge AI
Related Work
Cocoa AI Classification & RAG