AI-generated image detection on Stable Diffusion generator v1.5 (test)
92.3AccuracySAFE
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SAFETraining Dataset=ProGAN2026.01 | 92.3 | 98.83 | |
| Ours (ViT-L/14)Training Dataset=ProGAN, Backbone=ViT-L/142026.01 | 87.21 | 97.05 | |
| FatFormerTraining Dataset=ProGAN2026.01 | 83.9 | 96.24 | |
| NPRTraining Dataset=ProGAN2026.01 | 82.75 | 90.38 | |
| Ours (ViT-B/32)Training Dataset=ProGAN, Backbone=ViT-B/322026.01 | 82.75 | 93.87 | |
| GramNetTraining Dataset=ProGAN2026.01 | 65.15 | 62.86 | |
| LGradTraining Dataset=ProGAN2026.01 | 64.2 | 64.12 | |
| LNPTraining Dataset=ProGAN2026.01 | 61.1 | 68.23 | |
| CNNSpotTraining Dataset=ProGAN2026.01 | 52.2 | 58.06 | |
| UnivFDTraining Dataset=ProGAN2026.01 | 51.95 | 65.82 | |
| FreDectTraining Dataset=ProGAN2026.01 | 39.1 | 38.37 |