nii-yamagishilab
/

hubert-xlarge-anti-deepfake-nda

Model card Files Files and versions

nii-yamagishilab commited on Oct 21

Commit

8d248e6

·

verified ·

1 Parent(s): 96b086b

Update README.md

Files changed (1) hide show

README.md +1 -2

README.md CHANGED Viewed

@@ -200,8 +200,7 @@ Results shown below can be reproduced using scripts provided in our [GitHub repo
 | FakeOrReal           | 0.994   | 0.826    | 0.999     | 0.644  | 0.783    | 0.001 | 0.356 | 3.67 @ 0.0336        |
 | FakeOrReal-norm      | 0.910   | 0.784    | 0.991     | 0.563  | 0.718    | 0.005 | 0.437 | 15.56 @ 0.0758       |
 | In-the-Wild          | 0.919   | 0.733    | 0.969     | 0.594  | 0.737    | 0.032 | 0.406 | 17.99 @ 0.0841       |
-| Deepfake-Eval-2024  | 0.507   | 0.560    | 0.623     | 0.744  | 0.678    | 0.743 | 0.257 | 50.99 @ 0.8029       |
 You can also fine-tune this model on a specific database, the corresponding code is provided in our [GitHub repository](https://github.com/nii-yamagishilab/AntiDeepfake). Fine-tuning will follow a similar process to training a new model, except that model weights will be initialized as AntiDeepfake checkpoints.

 | FakeOrReal           | 0.994   | 0.826    | 0.999     | 0.644  | 0.783    | 0.001 | 0.356 | 3.67 @ 0.0336        |
 | FakeOrReal-norm      | 0.910   | 0.784    | 0.991     | 0.563  | 0.718    | 0.005 | 0.437 | 15.56 @ 0.0758       |
 | In-the-Wild          | 0.919   | 0.733    | 0.969     | 0.594  | 0.737    | 0.032 | 0.406 | 17.99 @ 0.0841       |
+| Deepfake-Eval-2024  | 0.541   | 0.580    | 0.668     | 0.711  | 0.689    | 0.669 | 0.289 | 47.73 @ 0.7125       |
 You can also fine-tune this model on a specific database, the corresponding code is provided in our [GitHub repository](https://github.com/nii-yamagishilab/AntiDeepfake). Fine-tuning will follow a similar process to training a new model, except that model weights will be initialized as AntiDeepfake checkpoints.