Object Counting On Carpk
Métriques
MAE
RMSE
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | MAE | RMSE |
|---|---|---|
| you-only-look-once-unified-real-time-object | 156.00 | 57.55 |
| faster-r-cnn-towards-real-time-object | 39.88 | 47.67 |
| a-large-contextual-dataset-for-classification | 21.88 | 36.73 |
| precise-detection-in-densely-packed-scenes | 6.77 | 8.52 |
| drone-based-object-counting-by-spatially | 22.76 | 34.46 |
| an-accurate-car-counting-in-aerial-images | 2.12 | 3.02 |
| vlcounter-text-aware-visual-representation | 6.46 | 8.68 |
| countr-transformer-based-generalised-visual | 5.75 | 7.45 |
| iterative-correlation-based-feature | 5.33 | 7.04 |
| yolo9000-better-faster-stronger | 130.40 | 172.46 |
| open-world-text-specified-object-counting | 8.13 | 10.87 |
| drone-based-object-counting-by-spatially | 16.62 | 22.30 |
| focal-loss-for-dense-object-detection | 24.58 | - |
| represent-compare-and-learn-a-similarity | 5.76 | 7.83 |