Object Detection on OV-LVIS
39.5AP (Rare)OV-DQUO
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| OV-DQUOBackbone=ViT-L/142026.07 | 39.5 | 33.7 | — | — | |
| CLIPSelf + F-ViT + ProCalBackbone=ViT-L/142026.07 | 39.3 | 37.3 | — | — | |
| CLIPSelf + NoOVDBackbone=ViT-L/142026.07 | 37.2 | 37.7 | — | — | |
| CLIPSelf + F-ViTBackbone=ViT-L/142026.07 | 36.8 | 37.3 | — | — | |
| OWL-ViTBackbone=ViT-L/142026.07 | 31.2 | 34.6 | — | — | |
| OV-DQUOBackbone=RN50x42026.07 | 29.4 | 26.5 | — | — | |
| CLIPSelf + NoOVDBackbone=RN50x42026.07 | 28.3 | 27.6 | — | — | |
| CORA+Backbone=RN50x42026.07 | 28.1 | — | — | — | |
| CLIPSelf + F-ViT + ProCalBackbone=RN50x42026.07 | 26.9 | 26 | — | — | |
| CoT-PL2025.10 | 26.4 | 34.9 | 34.8 | 38.2 | |
| F-VLMBackbone=RN50x42026.07 | 26.3 | 28.5 | — | — | |
| DST-DetBackbone=RN50x42026.07 | 26.2 | — | — | — | |
| BIRDet2025.10 | 26 | 25.5 | 21.7 | 29.5 | |
| CLIPSelf + F-ViTBackbone=RN50x42026.07 | 25.8 | 26.4 | — | — | |
| CAKE2025.10 | 25 | 34.9 | 34.8 | 38.4 | |
| DeticBackbone=RN502026.07 | 24.9 | 32.4 | — | — | |
| LBP2025.10 | 24.1 | 29.9 | 29.5 | 32.8 | |
| CoDet2025.10 | 23.4 | 30.7 | 30 | 34.6 | |
| BARON2025.10 | 23.2 | 29.5 | 29.3 | 32.5 | |
| MIC2025.10 | 22.9 | 34.4 | 34 | 39.9 | |
| DK-DETR2025.10 | 22.2 | 33.5 | 32 | 40.2 | |
| CORABackbone=RN50x42026.07 | 22.2 | — | — | — | |
| RegionCLIPBackbone=RN50x42026.07 | 22 | 32.3 | — | — | |
| OADP2025.10 | 21.9 | 28.7 | 28.4 | 32 | |
| RALF2025.10 | 21.9 | 26.6 | 26.2 | 29.1 | |
| OC-OVD2025.10 | 21.1 | 25.9 | 25 | 29.1 | |
| DetPro2025.10 | 20.8 | 28.4 | 27.8 | 32.4 | |
| F-VLMBackbone=RN502026.07 | 18.6 | 24.2 | — | — | |
| CCKT-Det++2025.10 | 18.2 | 27.1 | — | — | |
| RegionCLIP2025.10 | 17.1 | 28.2 | 27.4 | 34 | |
| RegionCLIPBackbone=RN502026.07 | 17.1 | 28.2 | — | — | |
| ViLD2025.10 | 16.7 | 27.8 | 26.5 | 34.2 | |
| DINO + ProvaBackbone=Swin-Base, Backbone Size=88M, Dbackbone=ImageNet-22K, V-L Pretraining=false2024.12 | 0.315 | 0.442 | — | — | |
| CFM-VITBackbone=ViT-B/16, Backbone Size=86M, Dbackbone=ALIGN, V-L Pretraining=true2024.12 | 0.296 | 0.338 | — | — | |
| RO-VITBackbone=ViT-B/16, Backbone Size=86M, Dbackbone=ALIGN, V-L Pretraining=false2024.12 | 0.284 | 0.319 | — | — | |
| CLIPSelfBackbone=ViT-B/16, Backbone Size=86M, Dbackbone=Merged-2B, V-L Pretraining=true2024.12 | 0.253 | — | — | — | |
| CORABackbone=ResNet-50x4, Backbone Size=87M, Dbackbone=CLIP-400M, V-L Pretraining=false2024.12 | 0.222 | — | — | — | |
| RegionCLIPBackbone=ResNet-50x4, Backbone Size=87M, Dbackbone=CLIP-400M, V-L Pretraining=true2024.12 | 0.22 | 0.323 | — | — |