Wavelet-Driven Generalizable Framework for Deepfake Face Forgery Detection
Lalith Bharadwaj Baru, Rohit Boddeda, Shilhora Akshay Patel, Sai Mohan Gajapaka
Abstract
The evolution of digital image manipulation, particu-larly with the advancement of deep generative models, sig-nificantly challenges existing deepfake detection methods, especially when the origin of the deepfake is obscure. To tackle the increasing complexity of these forgeries, we propose Wavelet-CLIP, a deepfake detection framework that integrates wavelet transforms with features derived from the ViT-L/14 architecture, pre-trained in the CLIP fash-ion. Wavelet-CLIP utilizes Wavelet Transforms to deeply analyze both spatial and frequency features from images, thus enhancing the model's capability to detect sophisti-cated deepfakes. To verify the effectiveness of our approach, we conducted extensive evaluations against existing state-of-the-art methods for cross-dataset generalization and de-tection of unseen images generated by standard diffusion models. Our method showcases outstanding performance, achieving an average AUC of 0.749 for cross-data generalization and 0.893 for robustness against unseen deep-fakes, outperforming all compared methods. The code can be reproduced from the repo: https://github.com/lalithbharadwajbaru/wavelet-clip