Image Captioning on ROCO v2
22.45METEOROmniRad base
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| OmniRad basebeams=5, max_tokens=1282026.02 | 22.45 | 2.95 | 19.31 | 53 | 17.47 | |
| OmniRad basebeams=5, max_tokens=642026.02 | 22.33 | 2.97 | 19.39 | 54 | 17.48 | |
| OmniRad basebeams=4, max_tokens=642026.02 | 22.25 | 2.93 | 19.18 | 53 | 17.3 | |
| DINOv3 basebeams=5, max_tokens=1282026.02 | 22.21 | 2.87 | 18.49 | 53 | 16.82 | |
| DINOv2 basebeams=5, max_tokens=642026.02 | 22.1 | 2.93 | 18.8 | 54 | 16.9 | |
| DINOv2 basebeams=5, max_tokens=1282026.02 | 22 | 2.9 | 18.8 | 53 | 16.95 | |
| DINOv3 basebeams=5, max_tokens=642026.02 | 21.99 | 2.91 | 18.82 | 54 | 16.96 | |
| DINOv3 basebeams=4, max_tokens=642026.02 | 21.97 | 2.85 | 18.66 | 51 | 16.9 | |
| DINOv2 basebeams=4, max_tokens=642026.02 | 21.9 | 2.87 | 18.7 | 52 | 16.9 | |
| CLIP basebeams=5, max_tokens=1282026.02 | 21.76 | 2.87 | 18.45 | 54 | 16.65 | |
| CLIP basebeams=5, max_tokens=642026.02 | 21.57 | 2.89 | 18.64 | 55 | 16.73 | |
| CLIP basebeams=4, max_tokens=642026.02 | 21.51 | 2.82 | 18.42 | 52 | 16.63 | |
| Radio DINO basebeams=4, max_tokens=642026.02 | 21.26 | 2.65 | 18.3 | 46 | 16.38 | |
| Radio DINO basebeams=5, max_tokens=642026.02 | 21.25 | 2.68 | 18.42 | 47 | 16.49 | |
| Radio DINO basebeams=5, max_tokens=1282026.02 | 21.25 | 2.68 | 18.42 | 47 | 16.49 | |
| OmniRad basebeams=1, max_tokens=642026.02 | 20.42 | 2.57 | 17.4 | 44 | 15.98 | |
| DINOv2 basebeams=1, max_tokens=642026.02 | 20.1 | 2.49 | 17.1 | 42 | 15.82 | |
| DINOv3 basebeams=1, max_tokens=642026.02 | 19.99 | 2.31 | 16.9 | 35 | 15.52 | |
| CLIP basebeams=1, max_tokens=642026.02 | 19.59 | 2.43 | 16.81 | 41 | 15.43 | |
| Radio DINO basebeams=1, max_tokens=642026.02 | 18.77 | 2.08 | 16.29 | 30 | 15.14 |