Table 2 Performance comparison of YOLOv8 and its variations with different IoU loss functions.
From: Attention-enhanced StrongSORT for robust vehicle tracking in complex environments
Method | Precision | Recall | mAP50 | FPS |
---|---|---|---|---|
YOLOv8 | 0.629 | 0.474 | 0.520 | 303 |
YOLOv8+GIoU | 0.612 | 0.480 | 0.531 | 303 |
YOLOv8+DIoU | 0.610 | 0.491 | 0.529 | 357 |
YOLOv8+CIoU | 0.656 | 0.479 | 0.532 | 357 |
YOLOv8+SIoU | 0.624 | 0.503 | 0.535 | 344 |
YOLOv8+EIoU | 0.621 | 0.483 | 0.534 | 294 |
YOLOv8+WIoU v1 | 0.590 | 0.481 | 0.530 | 303 |
YOLOv8+F-EIoU | 0.670 | 0.489 | 0.537 | 312 |