AI-generated image detection on ProGAN generator (test)
1AccuracyFatFormer
Evaluation results
| Method | Links | ||
|---|---|---|---|
| FatFormerTraining Dataset=ProGAN2026.01 | 1 | 1 | |
| Ours (ViT-L/14)Training Dataset=ProGAN, Backbone=ViT-L/142026.01 | 1 | 1 | |
| GramNetTraining Dataset=ProGAN2026.01 | 0.9999 | 1 | |
| Ours (ViT-B/32)Training Dataset=ProGAN, Backbone=ViT-B/322026.01 | 0.9999 | 1 | |
| SAFETraining Dataset=ProGAN2026.01 | 0.9998 | 1 | |
| NPRTraining Dataset=ProGAN2026.01 | 0.9996 | 1 | |
| UnivFDTraining Dataset=ProGAN2026.01 | 0.9981 | 1 | |
| LGradTraining Dataset=ProGAN2026.01 | 0.9978 | 1 | |
| CNNSpotTraining Dataset=ProGAN2026.01 | 0.9974 | 1 | |
| LNPTraining Dataset=ProGAN2026.01 | 0.9971 | 1 | |
| FreDectTraining Dataset=ProGAN2026.01 | 0.9936 | 0.9999 |