Object Detection on Golden 50 training samples regime (test)
70.5mAP50YOLO Ours
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| YOLO OursBackbone=YOLO, Training Strategy=Auto-labeled pretrain + Golden fine-tune2026.05 | 70.5 | 75.3 | 66.7 | 70.7 | 51.3 | |
| F.Rcnn OursBackbone=Faster R-CNN, Training Strategy=Auto-labeled pretrain + Golden fine-tune2026.05 | 70.2 | 46 | 75 | 57 | 42.9 | |
| F.Rcnn AutoBackbone=Faster R-CNN, Training Strategy=Direct training on auto-labeled dataset2026.05 | 60 | 19.3 | 86.9 | 31.6 | 39.5 | |
| F.Rcnn Golden +ECPBackbone=Faster R-CNN, Training Strategy=Direct training on Golden dataset, Hyperparameter Tuning=ECP2026.05 | 53.7 | 24.2 | 72.6 | 36.3 | 19.4 | |
| YOLO AutoBackbone=YOLO, Training Strategy=Direct training on auto-labeled dataset2026.05 | 46.4 | 47.8 | 51.2 | 49.4 | 14.4 | |
| Yolo Golden +ECPBackbone=YOLO, Training Strategy=Direct training on Golden dataset, Hyperparameter Tuning=ECP2026.05 | 30.2 | 28.9 | 44 | 34.9 | 14.4 |