Add pipeline tag, decision-model tag, and links to paper and code

#1
by nielsr HF Staff - opened
Files changed (1) hide show
  1. README.md +8 -1
README.md CHANGED
@@ -1,9 +1,16 @@
1
  ---
2
  license: cc-by-nc-4.0
 
 
 
3
  ---
4
 
5
  # TwinD: Generalizing Deepfake Detection from Blending to Entire Face Synthesis
6
 
 
 
 
 
7
  ![Overview of TwinD: reconstruction and blob twins of a real frame, training pairs of a fake and its real frame, and the mean AUROC on EFS and blending datasets](https://huggingface.co/yermandy/TwinD/resolve/main/fig/overview.png)
8
 
9
  Video-level AUROC (%) of SOTA methods on 19 datasets sorted by their macro-average (Avg). Above the dataset names are
@@ -30,4 +37,4 @@ their release years. † A temporal model, which takes several frames as input i
30
  | FS-VFM | 97.8 | 80.7 | 94.4 | 92.0 | 97.4 | 98.4 | 90.8 | 94.7 | 92.3 | 97.3 | 85.0 | 79.5 | 74.6 | 77.7 | 61.9 | 67.3 | 81.7 | 60.1 | 57.7 | 83.2 |
31
  | DFD-HR | 98.7 | 92.9 | 94.2 | 98.0 | 98.7 | 99.7 | 94.6 | 98.7 | 90.7 | 96.3 | 81.6 | 83.2 | 67.5 | 59.6 | 81.2 | 74.8 | 86.3 | 80.8 | 63.4 | 86.4 |
32
  | GenD-PE | 98.6 | 88.3 | 95.9 | 95.4 | 97.5 | 98.8 | 94.3 | 98.0 | 89.8 | 93.7 | 89.9 | 78.5 | 76.6 | 75.3 | 83.7 | 72.3 | 85.1 | 79.8 | 69.7 | 87.4 |
33
- | **TwinD** | **98.9** | **96.8** | **97.1** | **98.9** | **99.9** | **99.9** | **97.3** | **99.5** | **93.5** | **98.7** | 90.4 | **92.1** | **92.3** | 96.1 | **98.5** | **94.9** | **93.7** | **95.0** | **95.8** | **96.3** |
 
1
  ---
2
  license: cc-by-nc-4.0
3
+ pipeline_tag: image-classification
4
+ tags:
5
+ - decision-model
6
  ---
7
 
8
  # TwinD: Generalizing Deepfake Detection from Blending to Entire Face Synthesis
9
 
10
+ This repository contains the TwinD model presented in [TwinD: Generalizing Deepfake Detection from Blending to Entire Face Synthesis](https://huggingface.co/papers/2611.25935).
11
+
12
+ Code: https://github.com/yermandy/TwinD.
13
+
14
  ![Overview of TwinD: reconstruction and blob twins of a real frame, training pairs of a fake and its real frame, and the mean AUROC on EFS and blending datasets](https://huggingface.co/yermandy/TwinD/resolve/main/fig/overview.png)
15
 
16
  Video-level AUROC (%) of SOTA methods on 19 datasets sorted by their macro-average (Avg). Above the dataset names are
 
37
  | FS-VFM | 97.8 | 80.7 | 94.4 | 92.0 | 97.4 | 98.4 | 90.8 | 94.7 | 92.3 | 97.3 | 85.0 | 79.5 | 74.6 | 77.7 | 61.9 | 67.3 | 81.7 | 60.1 | 57.7 | 83.2 |
38
  | DFD-HR | 98.7 | 92.9 | 94.2 | 98.0 | 98.7 | 99.7 | 94.6 | 98.7 | 90.7 | 96.3 | 81.6 | 83.2 | 67.5 | 59.6 | 81.2 | 74.8 | 86.3 | 80.8 | 63.4 | 86.4 |
39
  | GenD-PE | 98.6 | 88.3 | 95.9 | 95.4 | 97.5 | 98.8 | 94.3 | 98.0 | 89.8 | 93.7 | 89.9 | 78.5 | 76.6 | 75.3 | 83.7 | 72.3 | 85.1 | 79.8 | 69.7 | 87.4 |
40
+ | **TwinD** | **98.9** | **96.8** | **97.1** | **98.9** | **99.9** | **99.9** | **97.3** | **99.5** | **93.5** | **98.7** | 90.4 | **92.1** | **92.3** | 96.1 | **98.5** | **94.9** | **93.7** | **95.0** | **95.8** | **96.3** |