Image Classification on CelebA Private Public
93.52AccuracyNo Defense
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| No DefenseAttack Method=KEDMI, Target Model (T)=IR152, Pre-trained Dataset (D_pretrain)=MS-CelebA-1M, |theta_c|/|theta_T|=62.6/62.62024.05 | 93.52 | 94.07 | 99.67 | 1,071 | |
| No DefenseAttack Method=GMI, Target Model (T)=IR152, Pre-trained Dataset (D_pretrain)=MS-CelebA-1M, |theta_c|/|theta_T|=62.6/62.62024.05 | 93.52 | 40.87 | 66.67 | 1,516 | |
| No DefenseAttack Method=KEDMI, Target Model (T)=VGG16, Pre-trained Dataset (D_pretrain)=ImageNet1K, |theta_c|/|theta_T|=16.8/16.82024.05 | 89 | 90.87 | 99.33 | 1,168 | |
| No DefenseAttack Method=GMI, Target Model (T)=VGG16, Pre-trained Dataset (D_pretrain)=ImageNet1K, |theta_c|/|theta_T|=16.8/16.82024.05 | 89 | 30.2 | 55 | 1,600 | |
| No DefenseAttack Method=KEDMI, Target Model (T)=FaceNet64, Pre-trained Dataset (D_pretrain)=MS-CelebA-1M, |theta_c|/|theta_T|=35.4/35.42024.05 | 88.5 | 86.73 | 98.33 | 1,194 | |
| No DefenseAttack Method=GMI, Target Model (T)=FaceNet64, Pre-trained Dataset (D_pretrain)=MS-CelebA-1M, |theta_c|/|theta_T|=35.4/35.42024.05 | 88.5 | 26.87 | 49 | 1,643 | |
| TL-DMIAttack Method=KEDMI, Target Model (T)=IR152, Pre-trained Dataset (D_pretrain)=MS-CelebA-1M, |theta_c|/|theta_T|=17.8/62.62024.05 | 86.7 | 64.6 | 87.67 | 1,333 | |
| TL-DMIAttack Method=GMI, Target Model (T)=IR152, Pre-trained Dataset (D_pretrain)=MS-CelebA-1M, |theta_c|/|theta_T|=17.8/62.62024.05 | 86.7 | 8.93 | 22.67 | 1,819 | |
| TL-DMIAttack Method=GMI, Target Model (T)=FaceNet64, Pre-trained Dataset (D_pretrain)=MS-CelebA-1M, |theta_c|/|theta_T|=34.4/35.42024.05 | 83.61 | 15.73 | 33 | 1,752 | |
| TL-DMIAttack Method=KEDMI, Target Model (T)=VGG16, Pre-trained Dataset (D_pretrain)=ImageNet1K, |theta_c|/|theta_T|=13.9/16.82024.05 | 83.41 | 51.67 | 80.33 | 1,410 | |
| TL-DMIAttack Method=KEDMI, Target Model (T)=FaceNet64, Pre-trained Dataset (D_pretrain)=MS-CelebA-1M, |theta_c|/|theta_T|=34.4/35.42024.05 | 83.41 | 73.4 | 91.67 | 1,265 | |
| TL-DMIAttack Method=GMI, Target Model (T)=VGG16, Pre-trained Dataset (D_pretrain)=ImageNet1K, |theta_c|/|theta_T|=13.9/16.82024.05 | 83.41 | 7.8 | 23.33 | 1,845 |