Object Detection on Watercolor
63.2mAPMR-DCoT
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MR-DCoTBackbone=Swin-T2026.07 | 63.2 | — | — | — | |
| SE-COTBackbone=Swin-T2026.07 | 60.7 | — | — | — | |
| MR-DCoTBackbone=Res1012026.07 | 60.3 | — | — | — | |
| SE-COTBackbone=Res1012026.07 | 57.5 | — | — | — | |
| DIV2026.07 | 57.4 | — | — | — | |
| NP2026.07 | 53.3 | — | — | — | |
| C-Gap2026.07 | 50.7 | — | — | — | |
| Faster R-CNN2026.07 | 45.6 | — | — | — | |
| VLOD-TTABackbone=Grounding DINO-B2025.10 | 44.7 | 72.8 | — | 46.9 | |
| ZSBackbone=Grounding DINO-B2025.10 | 42.6 | 70.5 | — | 44.6 | |
| VLOD-TTADetector=Grounding DINO2025.10 | 38.9 | 64.7 | — | 39.5 | |
| AdapterDetector=Grounding DINO2025.10 | 38.4 | 63.6 | — | 39 | |
| DPEDetector=Grounding DINO2025.10 | 37.6 | 63.2 | — | 38.1 | |
| ZSDetector=Grounding DINO2025.10 | 37.4 | 62.9 | — | 37.6 | |
| TPTDetector=Grounding DINO2025.10 | 37.4 | 63.1 | — | 37.6 | |
| VPTDetector=Grounding DINO2025.10 | 37.2 | 63 | — | 37.8 | |
| VLOD-TTABackbone=YOLO-World-L2025.10 | 34.1 | 58.3 | — | 34.2 | |
| ZSBackbone=YOLO-World-L2025.10 | 32.8 | 55.3 | — | 33 | |
| VLOD-TTADetector=YOLO-World2025.10 | 29.6 | 53.1 | — | 28.7 | |
| AdapterDetector=YOLO-World2025.10 | 28.3 | 51.5 | — | 26.7 | |
| TPTDetector=YOLO-World2025.10 | 27.3 | 48.5 | — | 26.1 | |
| DPEDetector=YOLO-World2025.10 | 27.2 | 48.9 | — | 26.3 | |
| ZSDetector=YOLO-World2025.10 | 26.9 | 47.9 | — | 25.9 | |
| VPTDetector=YOLO-World2025.10 | 26.9 | 49.1 | — | 25.1 | |
| CutLERBackbone=ResNet-50, Zero-shot=true, Unsupervised=true, Class-agnostic=true2023.01 | — | 37.5 | 44.6 | — | |
| FreeSOLOBackbone=ResNet-101, Zero-shot=true, Unsupervised=true, Class-agnostic=true2023.01 | — | 6.7 | 16.2 | — |