DeepFake Detection Using DenseNet-121
Amruthendu S Menon, Rasmi Raju, K S Lakshmi Varma, G S Anisha · 2025
In today’s world, image manipulations using advanced tools have become a significant concern. This paper introduces DenseNet-121 for distinguishing between real and fake images. The approach involves data preprocessing, organizing the dataset using CSV metadata, applying data augmentation, and training a fine-tuned convolutional neural network (CNN) to ensure high classification accuracy. Our implementation achieved a test accuracy of 97.00%.