ED$^4$: Explicit Data-level Debiasing for Deepfake Detection
원문 초록
Learning intrinsic bias from limited data has been considered the main reason for the failure of deepfake detection with generalizability. Apart from the discovered content and specific-forgery bias, we reveal a novel spatial bias, where detectors inertly anticipate observing structural forgery clues appearing at the image center, also can lead to the poor generalization of existing methods. We present ED$^4$, a simple and effective strategy, to address aforementioned biases explicitly at the data level in a unified framework rather than implicit disentanglement via network design. In particular, we develop ClockMix to produce facial structure preserved mixtures with arbitrary samples, which allows the detector to learn from an exponentially extended data distribution with much more diverse identities, backgrounds, local manipulation traces, and the co-occurrence of multiple forgery artifacts. We further propose the Adversarial Spatial Consistency Module (AdvSCM) to prevent extracting features with spatial bias, which adversarially generates spatial-inconsistent images and constrains their extracted feature to be consistent. As a model-agnostic debiasing strategy, ED$^4$ is plug-and-play: it can be integrated with various deepfake detectors to obtain significant benefits. We conduct extensive experiments to demonstrate its effectiveness and superiority over existing deepfake detection approaches.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
Cross-Branch Orthogonality for Improved Generalization in Face Deepfake Detection
Pay Less Attention to Deceptive Artifacts: Robust Detection of Compressed Deepfakes on Online Social Networks
FAME: A Lightweight Spatio-Temporal Network for Model Attribution of Face-Swap Deepfakes
Unmasking real-world audio deepfakes: A data-centric approach
TADA: Training-free Attribution and Out-of-Domain Detection of Audio Deepfakes
Multiverse Through Deepfakes: The MultiFakeVerse Dataset of Person-Centric Visual and Conceptual Manipulations
토론
이 논문과 연결된 포럼 글이 아직 없습니다.