Object Detection on OV-LVIS (val)
25.5AP (Novel)Ours
Evaluation results
| Method | Links | ||
|---|---|---|---|
| OursBackbone=ViT-B/16, Detector Architecture=F-ViT2026.04 | 25.5 | 24.7 | |
| OC-OVDBackbone=RN50, Detector Architecture=CenterNet2, Additional Supervision=IN-L, Additional Pretrained Models=MViT2026.04 | 25.2 | 32.9 | |
| CLIPSelfBackbone=ViT-B/16, Detector Architecture=F-ViT2026.04 | 25.1 | 24.5 | |
| DeticBackbone=RN50, Detector Architecture=CenterNet2, Additional Supervision=IN-L2026.04 | 24.6 | 32.4 | |
| CLIFFBackbone=RN50, Detector Architecture=CenterNet2, Additional Supervision=IN-L, Additional Pretrained Models=MViT2026.04 | 24.5 | 28.2 | |
| DV-DetBackbone=RN50, Detector Architecture=CenterNet2, Additional Supervision=CC3M2026.04 | 23.1 | 31.2 | |
| OursBackbone=RN50, Detector Architecture=Mask R-CNN2026.04 | 21.7 | 26 | |
| VL-DetBackbone=RN50, Detector Architecture=CenterNet2, Additional Supervision=CC3M2026.04 | 21.7 | 30.1 | |
| DK-DETRBackbone=RN50, Detector Architecture=DETR2026.04 | 20.5 | 30 | |
| OADPBackbone=RN50, Detector Architecture=Mask R-CNN2026.04 | 19.9 | — | |
| BARONBackbone=RN50, Detector Architecture=Mask R-CNN2026.04 | 19.2 | 26.5 | |
| F-VLMBackbone=RN50, Detector Architecture=Mask R-CNN2026.04 | 18.6 | 24.2 | |
| OV-DETRBackbone=RN50, Detector Architecture=DETR2026.04 | 17.4 | 26.6 |