MFCLIP: Multi-modal Fine-grained CLIP for Generalizable Diffusion Face Forgery Detection
About
The rapid development of photo-realistic face generation methods has raised significant concerns in society and academia, highlighting the urgent need for robust and generalizable face forgery detection (FFD) techniques. Although existing approaches mainly capture face forgery patterns using image modality, other modalities like fine-grained noises and texts are not fully explored, which limits the generalization capability of the model. In addition, most FFD methods tend to identify facial images generated by GAN, but struggle to detect unseen diffusion-synthesized ones. To address the limitations, we aim to leverage the cutting-edge foundation model, contrastive language-image pre-training (CLIP), to achieve generalizable diffusion face forgery detection (DFFD). In this paper, we propose a novel multi-modal fine-grained CLIP (MFCLIP) model, which mines comprehensive and fine-grained forgery traces across image-noise modalities via language-guided face forgery representation learning, to facilitate the advancement of DFFD. Specifically, we devise a fine-grained language encoder (FLE) that extracts fine global language features from hierarchical text prompts. We design a multi-modal vision encoder (MVE) to capture global image forgery embeddings as well as fine-grained noise forgery patterns extracted from the richest patch, and integrate them to mine general visual forgery traces. Moreover, we build an innovative plug-and-play sample pair attention (SPA) method to emphasize relevant negative pairs and suppress irrelevant ones, allowing cross-modality sample pairs to conduct more flexible alignment. Extensive experiments and visualizations show that our model outperforms the state of the arts on different settings like cross-generator, cross-forgery, and cross-dataset evaluations.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Attribution | Celeb-DF | Accuracy78.08 | 14 | |
| Attribution | WildDeepfake | Accuracy73.44 | 14 | |
| Attribution | Unseen Datasets Average | Accuracy60.69 | 14 | |
| Deepfake Attribution | DF40 and FFHQ unseen generators | FFHQ Accuracy14.48 | 14 | |
| Deepfake Detection | DF40 and FFHQ unseen generators | Average Accuracy (ACC)81.26 | 14 | |
| Deepfake Detection | Unseen Advanced Generators VAE, HART, FLUX | Average Accuracy37.94 | 14 | |
| Deepfake Attribution | Unseen Advanced Generators VAE, HART, FLUX | VAE Accuracy6.96 | 14 | |
| Attribution | DFDC | Accuracy30.56 | 14 | |
| Detection | Unseen Datasets Average | Accuracy53.08 | 14 | |
| Deepfake Attribution | Celeb-DF++ (unseen generators) | LivePortrait (ACC)80.56 | 9 |