Sports Ball Detection And Tracking On 2
Métriques
Accuracy (%)
Average Precision (%)
F1 (%)
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | Accuracy (%) | Average Precision (%) | F1 (%) | Paper Title | Repository |
|---|---|---|---|---|---|
| ResTrackNetV2 | 68.2 | 66.0 | 77.9 | Widely Applicable Strong Baseline for Sports Ball Detection and Tracking | |
| WASB (Step=3) | 71.3 | 71.5 | 80.6 | Widely Applicable Strong Baseline for Sports Ball Detection and Tracking | |
| DeepBall | 12.9 | 0 | 0 | DeepBall: Deep Neural-Network Ball Detector | |
| MonoTrack | 71.3 | 65.3 | 80.8 | MonoTrack: Shuttle trajectory reconstruction from monocular badminton video | |
| TrackNetV2 | 69.3 | 64.6 | 78.8 | TrackNetV2: Efficient Shuttlecock Tracking Network | |
| DeepBall-Large | 47.5 | 36.6 | 57.2 | Widely Applicable Strong Baseline for Sports Ball Detection and Tracking | |
| WASB (Step=1) | 73.4 | 77.1 | 82.6 | Widely Applicable Strong Baseline for Sports Ball Detection and Tracking | |
| BallSeg | 20.5 | 5.3 | 16.8 | Real-time CNN-based Segmentation Architecture for Ball Detection in a Single View Setup |
0 of 8 row(s) selected.