Towards Generalizable Deepfake Detection by Primary Region Regularization
Authors:
- Harry ChengHCHarry Cheng
- Yangyang GuoYGYangyang Guo
- Tianyi WangTWTianyi Wang
- Liqiang NieLNLiqiang Nie
- Mohan KankanhalliMKMohan Kankanhalli
Preprint
Abstract
The existing deepfake detection methods have reached a bottleneck in generalizing to unseen forgeries and manipulation approaches. Based on the observation that the deepfake detectors exhibit a preference for overfitting the specific primary regions in input, this paper enhances the generalization capability from a novel regularization perspective. This can be simply achieved by augmenting the images through primary region removal, thereby preventing the detector from over-relying on data bias. Our method consists of two stages, namely the static localization for primary region maps, as well as the dynamic exploitation of primary region masks. The proposed method can be seamlessly integrated into different backbones without affecting their inference efficiency. We conduct extensive experiments over three widely used deepfake datasets - DFDC, DF-1.0, and Celeb-DF with five backbones. Our method demonstrates an average performance improvement of 6% across different backbones and performs competitively with several state-of-the-art baselines.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Jointly Defending DeepFake Manipulation and Adversarial Attack Using Decoy MechanismIEEE Transactions on Pattern Analysis and Machine Intelligence, 2023
- ViTAEv2: Vision Transformer Advanced by Exploring Inductive Bias for Image Recognition and BeyondInternational Journal of Computer Vision, 2023
- Face Forgery Detection by 3D Decomposition and Composition SearchIEEE Transactions on Pattern Analysis and Machine Intelligence, 2023
- BoostTree and BoostForest for Ensemble LearningIEEE Transactions on Pattern Analysis and Machine Intelligence, 2022
- Supervised Anomaly Detection via Conditional Generative Adversarial Network and Ensemble Active LearningIEEE Transactions on Pattern Analysis and Machine Intelligence, 2022
- Obtaining genetics insights from deep learning via explainable artificial intelligenceNature Reviews Genetics, 2022
- FakeLocator: Robust Localization of GAN-Based Face ManipulationsIEEE Transactions on Information Forensics and Security, 2022
- Wasserstein Adversarial Regularization for Learning With Label NoiseIEEE Transactions on Pattern Analysis and Machine Intelligence, 2021
- DeepFake Detection Based on Discrepancies Between Faces and Their ContextIEEE Transactions on Pattern Analysis and Machine Intelligence, 2021
- Text-Guided Human Image Manipulation via Image-Text Shared SpaceIEEE Transactions on Pattern Analysis and Machine Intelligence, 2021