Object Detection on PASCAL VOC 19+1
77.1mAP@0.5 (1-19)CL-CLIP
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CL-CLIPBackbone=CLIP ViT-B/162026.06 | 77.1 | 84.8 | 77.5 | |
| NSGP-RePREBackbone=ResNet-502026.06 | 76.3 | 69 | 76 | |
| TARLBackbone=Swin-T2026.06 | 73.2 | 66.5 | 72.9 | |
| IODCBackbone=CLIP ViT-B/162026.06 | 72.8 | 62.5 | 72.3 | |
| OW-DETRBackbone=ResNet-502026.06 | 71.1 | 62 | 70.2 | |
| MMABackbone=ResNet-502026.06 | 71.1 | 63.4 | 70.7 | |
| ABRBackbone=ResNet-502026.06 | 71 | 69.7 | 70.9 | |
| GCDBackbone=Swin-T2026.06 | 69.8 | 74.6 | 70.1 | |
| ROSETTABackbone=ResNet-502026.06 | 69.7 | 68.3 | 69.6 | |
| BaselineBackbone=CLIP ViT-B/162026.06 | 68 | 74.3 | 68.3 |