Table 1 Detection performance of different IoU loss functions.
From: A deep learning based assisted analysis approach for Sjogren’s syndrome pathology images
Algorithm | P% | R% | mAP.5% | mAP.95% |
|---|---|---|---|---|
V8n+CIoU | 80.4 | 77 | 85.3 | 35.9 |
+DIOU | 78.6 | 84.7 | 87.8 | 36.8 |
+GIOU | 81.6 | 84 | 88.1 | 38 |
+EIOU | 77.1 | 82.3 | 85.7 | 34.6 |
+SIOU | 77.4 | 85.5 | 86.8 | 36.7 |
+MPDIOU | 77.7 | 85.6 | 88.4 | 37.7 |
+S-MPDIOU | 79.7 | 84.8 | 89 | 38 |
