- SSD SSD: Single Shot Multibox Detector
- FSSD FSSD: Feature Fusion Single Shot Multibox Detector
- RFB-SSDReceptive Field Block Net for Accurate and Fast Object Detection
- RefineDetSingle-Shot Refinement Neural Network for Object Detection
| System | mAP | FPS (Titan X Maxwell) | Our mAP |
|---|---|---|---|
| Faster R-CNN (VGG16) | 73.2 | 7 | |
| YOLOv2 (Darknet-19) | 78.6 | 40 | |
| R-FCN (ResNet-101) | 80.5 | 9 | |
| SSD300* (VGG16) | 77.2 | 46 | 77.6 |
| SSD512* (VGG16) | 79.8 | 19 | |
| RFBNet300 (VGG16) | 80.5 | 83 | |
| RFBNet512 (VGG16) | 82.2 | 38 | |
| FSSD300 (VGG) | 78.8 | 120 (1080Ti) | |
| FPN300 (VGG) | 78.3 |
| System | test-dev mAP | Time (Titan X Maxwell) | (0.5) | Our mAP |
|---|---|---|---|---|
| Faster R-CNN++ (ResNet-101)] | 34.9 | 3.36s | ||
| YOLOv2 (Darknet-19)] | 21.6 | 25ms | ||
| SSD300* (VGG16)] | 25.1 | 22ms | ||
| SSD512* (VGG16)] | 28.8 | 53ms | 43.1 | 42.9 |
| RetinaNet500 | 34.4 | 90ms | ||
| RFBNet300 (VGG16) | 29.9 | 15ms* | ||
| RFBNet512 (VGG16) | 33.8 | 30ms* | ||
| RFBNet512-E (VGG16) | 34.4 | 33ms* |