Object Detection on FishEye8K v1 (test)
38.4mAPAutomatic Text-guided Edge-case Synthesis (v1+v2)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Automatic Text-guided Edge-case Synthesis (v1+v2)Train dataset=train-D + naive v0 + automatic v1 + automatic v2, Count=50,992, Backbone=YOLOv11-small2025.09 | 38.4 | 36.6 | |
| Automatic Text-guided Edge-case Synthesis (v1)Train dataset=train-D + naive v0 + automatic v1, Count=35,057, Backbone=YOLOv11-small2025.09 | 38.1 | 36.3 | |
| YOLOv11-small (Naive Augmentation v0+v1+v2)Train dataset=train-D + naive v0 + naive v1 + naive v2, Count=50,992, Backbone=YOLOv11-small2025.09 | 38 | 36.2 | |
| YOLOv11-small (Manual Augmentation v1+v2)Train dataset=train-D + naive v0 + manual v1 + manual v2, Count=50,992, Backbone=YOLOv11-small2025.09 | 38 | 36.2 | |
| YOLOv11-small (Naive Augmentation v0+v1)Train dataset=train-D + naive v0 + naive v1, Count=35,057, Backbone=YOLOv11-small2025.09 | 37.8 | 36.1 | |
| YOLOv11-small (Naive Augmentation v0)Train dataset=train-D + naive v0, Count=19,122, Backbone=YOLOv11-small2025.09 | 37.6 | — | |
| YOLOv11-small (Manual Augmentation v1)Train dataset=train-D + naive v0 + manual v1, Count=35,057, Backbone=YOLOv11-small2025.09 | 37.4 | 35.7 | |
| YOLOv11-small (Baseline)Train dataset=train-D, Count=3,187, Backbone=YOLOv11-small2025.09 | 33.5 | — |