Multiview Detection On Wildtrack
평가 지표
MODA
MODP
Recall
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | MODA | MODP | Recall | Paper Title | Repository |
|---|---|---|---|---|---|
| MVDet | 88.2 | 75.7 | 93.6 | Multiview Detection with Feature Perspective Transformation | |
| MVDeTr | 91.5 | 82.1 | 94.0 | Multiview Detection with Shadow Transformer (and View-Coherent Data Augmentation) | |
| M-MVOT | 92.1 | - | - | Mahalanobis Distance-based Multi-view Optimal Transport for Multi-view Crowd Localization | |
| MVFlow | 91.9 | - | - | Multi-view Tracking Using Weakly Supervised Human Motion Prediction | |
| MVFP | 94.1 | 78.8 | 97.7 | Enhancing Multi-View Pedestrian Detection Through Generalized 3D Feature Pulling | - |
| EarlyBird | 91.2 | 81.8 | - | EarlyBird: Early-Fusion for Multi-View Tracking in the Bird's Eye View | |
| Deep-Occlusion | 74.1 | 53.8 | - | Deep Occlusion Reasoning for Multi-Camera Multi-Target Detection | |
| TrackTacular (Depth Splatting) | 93.2 | 77.5 | 95.8 | Lifting Multi-View Detection and Tracking to the Bird's Eye View | |
| 3DROM | 93.5 | 75.9 | 96.2 | 3D Random Occlusion and Multi-Layer Projection for Deep Multi-Camera Pedestrian Localization | |
| SHOT | 90.2 | 76.5 | 94.0 | Stacked Homography Transformations for Multi-View Pedestrian Detection | - |
0 of 10 row(s) selected.