GenImage
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| AI-generated image detection | GenImage | Midjourney Detection Rate99.54 | 173 | |
| Black-box Attack | GenImage | ASR99.9 | 162 | |
| Generated Image Detection | GenImage (test) | Average Accuracy99.7 | 135 | |
| AI-generated image detection | GenImage (test) | Mean Accuracy97.16 | 51 | |
| AI-generated image detection | GenImage 61 (test) | AUC96.9 | 45 | |
| Generated Image Detection | GenImage v1.4 (test) | Average AP100 | 34 | |
| Adversarial Attack Imperceptibility | GenImage | PSNR34.7 | 30 | |
| AI-Generated Image Detection | GenImage 1.0 (test) | Midjourney Detection Rate95.61 | 24 | |
| AI Image Detection | GenImage (OOD) | AUROC99.84 | 19 | |
| Synthetic Image Detection | GenImage (test) | Midjourney Performance96.33 | 18 | |
| AIGI Detection | GenImage v1.4 (test) | ADM Score0.88 | 18 | |
| Forgery Type Identification | GenImage | Accuracy91 | 17 | |
| AI-Generated Image Detection | GenImage v1.4 (test) | Midjourney Detection Accuracy98 | 15 | |
| Cross-domain AI-generated image detection | GenImage | F1 Score94.78 | 15 | |
| AI-Generated Image Detection | GenImage 1.0 (various) | Midjourney Accuracy100 | 14 | |
| AI-generated image detection | GenImage SD v1.4 March 2024 | Detection Accuracy (MJ)95.9 | 14 | |
| Synthetic Image Detection | GenImage | Detection Rate (Midjourney)97.6 | 12 | |
| Image-level AIGC Forgery Detection | GenImage | Accuracy99.7 | 12 | |
| AI-generated image attribution | GenImage | Rank-1 (Real)83.4 | 12 | |
| AI-generated image attribution | GenImage cross-generator setting (test) | Rank-1 (Real)89.4 | 12 | |
| Adversarial Attack | GenImage SD | CNNSpot Performance100 | 12 | |
| AIGI Detection | GenImage | B.Acc94.2 | 12 | |
| Fake Image Detection | GenImage++ (test) | Flux Score99.85 | 11 | |
| Synthetic Image Detection | GenImage | Balanced Accuracy98.2 | 11 | |
| Synthetic Image Detection | GenImage VQDM (test) | Accuracy98.9 | 11 |
Showing 25 of 53 rows