| Benchmark | Model | Rank | Results |
|---|---|---|---|
| pedestrian-detection-on-dvtod | YOLOv3 (Thermal) | #2 | mAP: 82.7 |
| pedestrian-detection-on-dvtod | YOLOv3 (Visible) | #6 | mAP: 34.5 |
| classification-on-indl | Darknet53 | #7 | Average Recall: 88.53% |
| robust-object-detection-on-cityscapes | Photometric distortion | #12 | mPC [AP]: 16.9 |
| real-time-object-detection-on-coco-test-dev | YOLOv3-608 | #22 | box AP: 33.0FPS (batch 1, GPU in brackets): 20 (Titan X) |
| object-detection-on-coco-o | YOLOv3 (DarkNet-53) | #41 | Average mAP: 14.8Effective Robustness: -0.37 |