Object Counting On Fsc147
Métriques
MAE(test)
MAE(val)
RMSE(test)
RMSE(val)
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | MAE(test) | MAE(val) | RMSE(test) | RMSE(val) |
|---|---|---|---|---|
| a-novel-unified-architecture-for-low-shot | 7.91 | 9.52 | 54.28 | 43.00 |
| learning-to-count-anything-reference-less | 17.12 | 17.49 | 104.53 | 58.81 |
| iterative-correlation-based-feature | 14.32 | 15.28 | 85.54 | 47.20 |
| few-shot-object-counting-and-detection | 16.79 | - | 123.56 | - |
| gca-sun-a-gated-context-aware-swin-unet-for | 14.00 | 16.06 | 92.19 | 53.04 |
| countr-transformer-based-generalised-visual | 11.95 | 13.13 | 91.23 | 49.83 |
| semantic-generative-augmentations-for-few | 12.74 | 12.59 | 89.90 | 44.95 |
| dave-a-detect-and-verify-paradigm-for-low | 8.66 | 8.91 | 32.36 | 28.08 |
| omnicount-multi-label-object-counting-with | 18.63 | - | 112 | - |
| learning-to-count-everything | 22.08 | 23.75 | 99.54 | 69.07 |
| vision-transformer-off-the-shelf-a-surprising | 9.13 | 10.63 | 48.96 | 37.95 |
| semantic-generative-augmentations-for-few | 11.32 | 12.31 | 77.50 | 41.65 |
| countgd-multi-modal-open-world-counting | 5.74 | 7.1 | 24.09 | 26.08 |
| object-counting-you-only-need-to-look-at-one | 15.78 | 17.11 | 97.15 | 56.81 |
| open-world-text-specified-object-counting | 15.88 | 17.10 | 106.29 | 65.61 |
| learning-spatial-similarity-distribution-for | 9.58 | 9.73 | 64.13 | 29.72 |
| scale-prior-deformable-convolution-for | 13.51 | 14.59 | 96.80 | 49.97 |
| a-low-shot-object-counting-network-with | 10.79 | 10.24 | 56.97 | 32.56 |
| represent-compare-and-learn-a-similarity | 14.62 | 15.74 | 91.83 | 58.53 |