Deep learning models compared for detecting fake facial images
Deep Learning for Deepfake Image Detection: Comparative Model Evaluation and Gradient-Based Visual Attribution
Computer Vision and Pattern Recognition
Summary
Distinguishing real faces from digitally faked faces is tough but important for online trust. This paper compares two deep learning methods: one smaller model trained from scratch and a larger model adapted from a general image database. The larger model was more accurate at spotting fakes. The authors used a visualization technique to show which facial areas the models focused on when deciding if an image was fake. They also created a simple tool to look at these explanations side by side.
What this means in practice
- •For digital security teams: Use pretrained deep learning models to improve detection of fake facial images in online content moderation.
- •For mobile app developers: Integrate compact deep learning models for on-device fake face detection with visualization to explain suspicious detections.
Tested on one dataset.
Authors
Saad Korchi, Abderrahim Waga
Abstract
Synthetic facial imagery has made the distinction between authentic and generated content increasingly important for digital forensics and online trust. This study examines an explainable deep-learning workflow for image-level deepfake detection using two contrasting classification strategies: a compact convolutional model trained from scratch and an ImageNet-pretrained MobileNetV2 model adapted through two-stage transfer learning. Experiments were conducted with the balanced 140k Real and Fake Faces dataset, comprising 140,000 facial images. Under the recorded validation protocol, the augmented from-scratch model reached 84.15% accuracy (loss 0.3625), compared with 95.90% accuracy (loss 0.1074) for the fine-tuned pretrained model. Gradient-weighted class activation mapping (Grad-CAM) was used to visualize image regions associated with predictions, and a Flask-based prototype was developed for side-by-side inspection. These observations support the practical value of pretrained visual representations in the studied setting while underscoring the need to interpret validation performance alongside dataset provenance, explanation limitations, and evaluation scope. The reported results are validation results, not held-out test estimates.