Full name | MesoNet: a Compact Facial Video Forgery Detection Network |
Description | Based on the corresponding paper, trained the same way as the Xception network as outline in the FaceForensics++ paper. Trained on all images compressed with h264 with a crf of 40.
Face Tracking: dlib |
Publication title | MesoNet: a Compact Facial Video Forgery Detection Network |
Publication authors | Darius Afchar, Vincent Nozick, Junichi Yamagishi, Isao Echizen |
Publication venue | WIFS 2018 |
Publication URL | https://arxiv.org/abs/1809.00888 |
Input Data Types | Uses Face Detection |
Programming language(s) | Python, Keras |
Hardware | GeForce 1080ti |
Submission creation date | 13 Mar, 2019 |
Last edited | 13 Mar, 2019 |