Table 2 Comparison table of detection algorithms for real datasets.
From: A deep learning based detection algorithm for anomalous behavior and anomalous item on buses
| Â | Time(fps) | mAP |
|---|---|---|
RetinaMask-101 [10] | 12 | 83.8% |
Faster R-CNN [12] | 8 | 88.0% |
YOLOv5 | 120 | 90.5% |
YOLOv7 | 149 | 89.1% |
YOLOv8 | 128 | 90.2% |
RT-DETR | 114 | 85.4% |
MD-AODA | 101 | 92.6% |