AI-Generated Image Detection on DRCT-2M
99.99LDM Detection RateCLIP/RN50
Evaluation Results
| Method | Links | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CLIP/RN50Training Dataset=SD v1.42025.11 | 99.99 | 99.99 | 99.96 | 94.3 | 38.94 | 90.63 | 80.34 | 47.74 | 98.96 | 25.9 | 99.97 | 76.07 | 78.1 | 48.11 | 2.68 | 1.9 | 67.72 | |
| Conv-BTraining Dataset=SD v1.42025.11 | 99.97 | 100 | 99.97 | 95.66 | 44.82 | 78.05 | 76.27 | 35.39 | 99.26 | 39.56 | 99.8 | 80.1 | 63.54 | 37.79 | 6.91 | 1.63 | 66.17 | |
| Conv-BTrained on=SD v1.42025.11 | 99.97 | 100 | 99.97 | 95.84 | 64.44 | 82 | 80.82 | 60.75 | 99.27 | 62.33 | 99.8 | 83.4 | 73.28 | 61.65 | 51.79 | 50.41 | 79.11 | |
| DRCTTraining Dataset=SD v1.42025.11 | 99.91 | 99.9 | 99.9 | 96.19 | 80.81 | 83.25 | 91.18 | 57.33 | 99.66 | 73.09 | 99.9 | 94.76 | 76.91 | 99.9 | 95.19 | 67.43 | 88.46 | |
| DRCTTrained on=SD v1.42025.11 | 99.91 | 99.9 | 99.9 | 96.32 | 83.87 | 85.63 | 91.88 | 70.04 | 99.66 | 78.76 | 99.9 | 95.01 | 81.21 | 99.9 | 95.4 | 75.39 | 90.79 | |
| CNNSpotTraining Dataset=SD v1.42025.11 | 99.87 | 99.91 | 99.9 | 97.49 | 49.13 | 84.48 | 83.94 | 61.97 | 98.23 | 38.08 | 97.92 | 83.59 | 79.31 | 35.98 | 5.67 | 1.31 | 69.8 | |
| CNNSpotTrained on=SD v1.42025.11 | 99.87 | 99.91 | 99.9 | 97.55 | 66.25 | 86.55 | 86.15 | 72.42 | 98.26 | 61.72 | 97.96 | 85.89 | 82.84 | 60.93 | 51.41 | 50.28 | 81.12 | |
| F3NetTraining Dataset=SD v1.42025.11 | 99.85 | 99.78 | 99.79 | 84.24 | 21.2 | 85.57 | 53.69 | 43.08 | 97.32 | 18.38 | 97.94 | 61.95 | 78.08 | 47.29 | 1.9 | 1.43 | 61.97 | |
| F3NetTrained on=SD v1.42025.11 | 99.85 | 99.78 | 99.79 | 88.66 | 55.85 | 87.37 | 68.29 | 63.66 | 97.39 | 54.98 | 97.98 | 72.39 | 81.99 | 65.42 | 50.39 | 50.27 | 77.13 | |
| GramNetTraining Dataset=SD v1.42025.11 | 99.4 | 99.01 | 98.83 | 95.1 | 40.99 | 76.26 | 59.92 | 56.18 | 92.59 | 25.54 | 88.94 | 67.93 | 79.23 | 6.09 | 1.42 | 1.69 | 61.82 | |
| GramNetTrained on=SD v1.42025.11 | 99.4 | 99.01 | 98.84 | 95.3 | 62.63 | 80.68 | 71.19 | 69.32 | 93.05 | 57.02 | 89.97 | 75.55 | 82.68 | 51.23 | 50.01 | 50.08 | 76.62 | |
| CLIP/RN50Trained on=SD v1.42025.11 | 99 | 99.99 | 99.96 | 94.61 | 62.08 | 91.43 | 83.57 | 64.4 | 98.97 | 57.43 | 99.74 | 80.69 | 82.03 | 65.83 | 50.67 | 50.47 | 80.05 | |
| UnivFDTrained on=SD v1.42025.11 | 98.3 | 96.22 | 96.33 | 93.83 | 91.01 | 93.91 | 86.38 | 85.92 | 90.44 | 88.99 | 90.41 | 81.06 | 89.06 | 51.96 | 51.03 | 50.46 | 83.46 | |
| UnivFDTraining Dataset=SD v1.42025.11 | 98.29 | 96.11 | 96.22 | 93.48 | 90.21 | 93.57 | 84.39 | 83.78 | 89.53 | 87.75 | 89.49 | 76.88 | 87.83 | 9.01 | 5.63 | 3.47 | 74.1 | |
| DIRETrained on=SD v1.42025.11 | 98.19 | 99.94 | 99.96 | 68.16 | 53.84 | 71.93 | 58.87 | 54.35 | 99.78 | 59.73 | 99.65 | 64.2 | 59.13 | 51.99 | 50.04 | 49.97 | 71.23 | |
| DIRETraining Dataset=SD v1.42025.11 | 98.16 | 99.94 | 99.96 | 53.33 | 14.36 | 61.01 | 30.21 | 16.1 | 99.78 | 32.65 | 99.65 | 44.29 | 30.95 | 7.76 | 0.28 | 0 | 49.28 | |
| De-fakeTrained on=SD v1.42025.11 | 92.1 | 99.53 | 99.51 | 89.65 | 64.02 | 69.24 | 92 | 93.93 | 99.13 | 70.89 | 58.98 | 62.34 | 66.66 | 50.12 | 50.16 | 50 | 75.52 | |
| De-fakeTraining Dataset=SD v1.42025.11 | 91.45 | 99.53 | 99.51 | 88.5 | 44.1 | 55.79 | 91.34 | 93.56 | 99.13 | 59.13 | 30.85 | 39.92 | 50.24 | 1.15 | 1.31 | 0.68 | 59.14 | |
| OmniAID-MirageTrained on=Mirage-Train2025.11 | 90.62 | 98.45 | 98.43 | 96.6 | 92.02 | 97.33 | 82.34 | 71.6 | 97.86 | 98.61 | 94.19 | 72.51 | 84.2 | 98.88 | 98.82 | 98.15 | 91.91 | |
| OmniAID-MirageTraining Dataset=Mirage-Train2025.11 | 89.9 | 98.46 | 98.44 | 96.56 | 91.53 | 97.32 | 79.12 | 61.53 | 97.86 | 98.62 | 93.97 | 63.21 | 81.72 | 98.89 | 98.83 | 98.16 | 90.26 |