Face Forgery Detection via Symmetric Transformer
Lists
Only you see this
Abstract
The deep learning-based face forgery detection is a novel yet challenging task. Despite impressive results have been achieved, thereare still some limitations in the existing methods. For example, the previous methods are hard to maintain consistent predictions for consecutive frames, even if all of those frames are actually forged. We propose a symmetric transformer for channel and spatial feature extraction, which is because the channel and spatial features of a robust forgery detector should be consistent in the temporal domain. The symmetric transformer adopt the newly-designed attentionbased strategies for channel variance and spatial gradients as the vital features, which greatly improves the robustness of deepfake video detection. Moreover, this symmetric structure acts on temporal and spatial features respectively, which ensures the robustness of detection from two different aspects. Our symmetric transformer is an end-to-end optimized network. Experiments are conducted on various settings, the proposed methods achieve significantly improvement on prediction robustness and perform better than state-of-the-art methods on different datasets.
BibTeX
@inproceedings{song2022face,
author = {Luchuan Song},
title = {{Face Forgery Detection via Symmetric Transformer}},
booktitle = {ACM Multimedia},
year = {2022},
doi = {10.1145/3503161.3547806},
}Loading related content...