Object Detection on Comic
40.4mAPMR-DCoT
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MR-DCoTBackbone=Swin-T2026.07 | 40.4 | — | — | — | |
| VLOD-TTABackbone=Grounding DINO-B2025.10 | 39.9 | 67.1 | — | 41.1 | |
| ZSBackbone=Grounding DINO-B2025.10 | 38.7 | 64.7 | — | 39.2 | |
| MR-DCoTBackbone=Res1012026.07 | 37.3 | — | — | — | |
| SE-COTBackbone=Swin-T2026.07 | 36.9 | — | — | — | |
| SE-COTBackbone=Res1012026.07 | 34.8 | — | — | — | |
| VLOD-TTADetector=Grounding DINO2025.10 | 34.2 | 57.8 | — | 35.3 | |
| DIV2026.07 | 33.2 | — | — | — | |
| DPEDetector=Grounding DINO2025.10 | 31.8 | 53.3 | — | 31.7 | |
| AdapterDetector=Grounding DINO2025.10 | 31.7 | 54.1 | — | 32 | |
| TPTDetector=Grounding DINO2025.10 | 31.5 | 53.6 | — | 31.8 | |
| ZSDetector=Grounding DINO2025.10 | 31.2 | 52.9 | — | 31.5 | |
| VPTDetector=Grounding DINO2025.10 | 31.1 | 52.6 | — | 31.2 | |
| C-Gap2026.07 | 29.4 | — | — | — | |
| NP2026.07 | 28.9 | — | — | — | |
| VLOD-TTABackbone=YOLO-World-L2025.10 | 26.5 | 42.8 | — | 27.2 | |
| ZSBackbone=YOLO-World-L2025.10 | 23.3 | 37.9 | — | 23.6 | |
| VLOD-TTADetector=YOLO-World2025.10 | 21.4 | 36.1 | — | 22.1 | |
| AdapterDetector=YOLO-World2025.10 | 20.8 | 34.7 | — | 21.7 | |
| Faster R-CNN2026.07 | 19.4 | — | — | — | |
| DPEDetector=YOLO-World2025.10 | 18.9 | 31.7 | — | 19.8 | |
| VPTDetector=YOLO-World2025.10 | 18.3 | 30.9 | — | 19.3 | |
| TPTDetector=YOLO-World2025.10 | 18.1 | 29.9 | — | 19.1 | |
| ZSDetector=YOLO-World2025.10 | 17.8 | 29.4 | — | 18.8 | |
| CutLERBackbone=ResNet-50, Zero-shot=true, Unsupervised=true, Class-agnostic=true2023.01 | — | 30.4 | 38.6 | — | |
| FreeSOLOBackbone=ResNet-101, Zero-shot=true, Unsupervised=true, Class-agnostic=true2023.01 | — | 9.9 | 16.3 | — |