-
[pdf]
[supp]
[bibtex]@InProceedings{Yao_2023_ICCV, author = {Yao, Kelu and Wang, Jin and Diao, Boyu and Li, Chao}, title = {Towards Understanding the Generalization of Deepfake Detectors from a Game-Theoretical View}, booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)}, month = {October}, year = {2023}, pages = {2031-2041} }
Towards Understanding the Generalization of Deepfake Detectors from a Game-Theoretical View
Abstract
This paper aims to explain the generalization of deepfake detectors from the novel perspective of multi-order interactions among visual concepts. Specifically, we propose three hypotheses:
1. Deepfake detectors encode multi-order interactions among visual concepts, in which the low-order interactions usually have substantially negative contributions to deepfake detection.
2. Deepfake detectors with better generalization abilities tend to encode low-order interactions with fewer negative contributions.
3. Generalized deepfake detectors usually weaken the negative contributions of low-order interactions by suppressing their strength.
Accordingly, we design several mathematical metrics to evaluate the effect of low-order interaction for deepfake detectors.
Extensive comparative experiments are conducted, which verify the soundness of our hypotheses.
Based on the analyses, we further propose a generic method, which directly reduces the toxic effects of low-order interactions to improve the generalization of deepfake detectors to some extent.
The code will be released when the paper is accepted.
Related Material