Corpus ID: 219688005

FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction

@article{Huang2020FakePolisherMD,
  title={FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction},
  author={Yihao Huang and Felix Juefei-Xu and Run Wang and Qing Guo and Lei Ma and Xiaofei Xie and Jianwen Li and Weikai Miao and Yang Liu and Geguang Pu},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.07533}
}
  • Yihao Huang, Felix Juefei-Xu, +7 authors Geguang Pu
  • Published 2020
  • Computer Science
  • ArXiv
  • The recently rapid advances of generative adversarial networks (GANs) in synthesizing realistic and natural DeepFake information (e.g., images, video) cause severe concerns and threats to our society. At this moment, GAN-based image generation methods are still imperfect, whose upsampling design has limitations in leaving some certain artifact patterns in the synthesized image. Such artifact patterns can be easily exploited (by recent methods) for difference detection of real and GAN… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-3 OF 3 CITATIONS

    DeepSonar: Towards Effective and Robust Detection of AI-Synthesized Fake Voices

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Pasadena: Perceptually Aware and Stealthy Adversarial Denoise Attack

    VIEW 1 EXCERPT
    CITES BACKGROUND

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 53 REFERENCES

    Large-scale celebfaces attributes (celeba) dataset

    • Ziwei Liu, Ping Luo, Xiaogang Wang, Xiaoou Tang
    • Retrieved August
    • 2018
    VIEW 10 EXCERPTS
    HIGHLY INFLUENTIAL

    A Style-Based Generator Architecture for Generative Adversarial Networks

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    StarGAN: Unified Generative Adversarial Networks for Multi-domain Image-to-Image Translation

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Analyzing and Improving the Image Quality of StyleGAN

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints

    • Ning Yu, Larry Davis, Mario Fritz
    • Computer Science
    • 2019 IEEE/CVF International Conference on Computer Vision (ICCV)
    • 2019
    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Detecting and Simulating Artifacts in GAN Fake Images

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    FaceForensics++: Learning to Detect Manipulated Facial Images

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL