Multiview Detection On Citystreet
Metrics
F1_score (2m)
MODA (2m)
MODP (2m)
Precision (2m)
Recall (2m)
Results
Performance results of various models on this benchmark
Model Name | F1_score (2m) | MODA (2m) | MODP (2m) | Precision (2m) | Recall (2m) | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| 3DROM | 79.2 | 60.0 | 70.1 | 82.5 | 76.2 | 3D Random Occlusion and Multi-Layer Projection for Deep Multi-Camera Pedestrian Localization | |
| MVDet | 68.4 | 44.6 | 65.7 | 79.8 | 59.8 | Multiview Detection with Feature Perspective Transformation | |
| SVCW | 76.0 | 55.0 | 70.0 | 81.4 | 71.2 | Multi-View People Detection in Large Scenes via Supervised View-Wise Contribution Weighting | |
| SHOT | 71.8 | 53.5 | 72.4 | 91.0 | 59.4 | Stacked Homography Transformations for Multi-View Pedestrian Detection | - |
| MVDeTr | 75.2 | 58.3 | 74.1 | 92.8 | 63.2 | Multiview Detection with Shadow Transformer (and View-Coherent Data Augmentation) |
0 of 5 row(s) selected.