Camouflaged Object Detection on OVCOD-D 1.0 (union of base and novel classes)
56.4APSDDF-L
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SDDF-LBackbone=YOLOv8-L, Params=109M, Pre-train=O365,GoldG2026.03 | 56.4 | 76.4 | 60.7 | 34.4 | 59 | |
| SDDF-MBackbone=YOLOv8-M, Params=91M, Pre-train=O365,GoldG2026.03 | 54.3 | 75.1 | 57.5 | 30.3 | 57 | |
| DOSOD-LBackbone=YOLOv8-L, Params=108M, Pre-train=O365,GoldG2026.03 | 53.4 | 73.1 | 56.2 | 26.4 | 56.3 | |
| DOSOD-MBackbone=YOLOv8-M, Params=90M, Pre-train=O365,GoldG2026.03 | 51.8 | 73.8 | 55.6 | 27.9 | 54.5 | |
| SDDF-SBackbone=YOLOv8-S, Params=76M, Pre-train=O365,GoldG2026.03 | 48.1 | 70.7 | 50.3 | 25.9 | 50.7 | |
| YOLO-World-LBackbone=YOLOv8-L, Params=110M, Pre-train=O365,GoldG2026.03 | 45.7 | 63.2 | 48.9 | 22.9 | 48.4 | |
| DOSOD-SBackbone=YOLOv8-S, Params=75M, Pre-train=O365,GoldG2026.03 | 44.8 | 67.4 | 46.8 | 22.4 | 47.2 | |
| YOLO-World-MBackbone=YOLOv8-M, Params=92M, Pre-train=O365,GoldG2026.03 | 43.3 | 62.3 | 46.5 | 24.9 | 46.1 | |
| YOLOE-MBackbone=YOLOv8-M, Params=94M, Pre-train=O365,GoldG2026.03 | 39.9 | 47.7 | 42.7 | — | — | |
| YOLOE-SBackbone=YOLOv8-S, Params=78M, Pre-train=O365,GoldG2026.03 | 38.7 | 46 | 40.8 | — | — | |
| YOLO-World-SBackbone=YOLOv8-S, Params=77M, Pre-train=O365,GoldG2026.03 | 38.5 | 58 | 41.4 | 18.7 | 41.1 | |
| Grounding DINO-TBackbone=Swin-T, Params=172M, Pre-train=O365,GoldG2026.03 | 34.8 | 43.9 | 37.7 | — | — | |
| GLIPv2-TBackbone=Swin-T, Params=232M, Pre-train=O365,GoldG2026.03 | 25.6 | 36.4 | 29.7 | — | — | |
| GLIP-TBackbone=Swin-T, Params=232M, Pre-train=O365,GoldG2026.03 | 22.6 | 31.9 | 28.2 | — | — |