Object Detection on COCO (AP)
68.4APCP-DETR-L*
Evaluation Results
| Method | Links | |
|---|---|---|
| CP-DETR-L*Open Source=N, Prompt Type=Visual-I, Backbone=Swin-L, VLM sup.=CLIP, Data Size=3.3M, Training Data=O365, OpenImages, GoldG, V3Det2026.04 | 68.4 | |
| PET-DINOOpen Source=Y, Prompt Type=Visual-I, Backbone=Swin-L, VLM sup.=-, Data Size=2.5M, Training Data=O365, OpenImages, V3Det2026.04 | 66.5 | |
| PET-DINOOpen Source=Y, Prompt Type=Visual-I, Backbone=Swin-L, VLM sup.=-, Data Size=0.6M, Training Data=O3652026.04 | 66.4 | |
| PET-DINOOpen Source=Y, Prompt Type=Visual-I, Backbone=Swin-T, VLM sup.=-, Data Size=2.5M, Training Data=O365, OpenImages, V3Det2026.04 | 64.3 | |
| PET-DINOOpen Source=Y, Prompt Type=Visual-I, Backbone=Swin-T, VLM sup.=-, Data Size=0.6M, Training Data=O3652026.04 | 64 | |
| CP-DETR-TOpen Source=N, Prompt Type=Visual-I, Backbone=Swin-T, VLM sup.=CLIP, Data Size=3.3M, Training Data=O365, OpenImages, GoldG, V3Det2026.04 | 61.8 | |
| T-Rex2Open Source=N, Prompt Type=Visual-I, Backbone=Swin-L, VLM sup.=CLIP, Data Size=3.1M, Training Data=O365, OpenImages, HierText, CrowdHuman, SA-1B2026.04 | 58.5 | |
| VL-SAM-v3Backbone=Swin-T2026.05 | 56.8 | |
| T-Rex2Open Source=N, Prompt Type=Visual-I, Backbone=Swin-T, VLM sup.=CLIP, Data Size=3.1M, Training Data=O365, OpenImages, HierText, CrowdHuman, SA-1B2026.04 | 56.6 | |
| VL-SAM-v2Backbone=Swin-T2026.05 | 56 | |
| LLMDetBackbone=Swin-T2026.05 | 55.6 | |
| MM-GDINOBackbone=Swin-T2026.05 | 50.4 | |
| Grounding-DINOBackbone=Swin-T2026.05 | 48.4 | |
| DetCLIPv3Backbone=Swin-T2026.05 | 47.2 | |
| PET-DINOOpen Source=Y, Prompt Type=Visual-G, Backbone=Swin-L, VLM sup.=-, Data Size=0.6M, Training Data=O3652026.04 | 46.7 | |
| T-Rex2Open Source=N, Prompt Type=Visual-G, Backbone=Swin-L, VLM sup.=CLIP, Data Size=3.1M, Training Data=O365, OpenImages, HierText, CrowdHuman, SA-1B2026.04 | 46.5 | |
| GLIPBackbone=Swin-T2026.05 | 46.1 | |
| PET-DINOOpen Source=Y, Prompt Type=Visual-G, Backbone=Swin-L, VLM sup.=-, Data Size=2.5M, Training Data=O365, OpenImages, V3Det2026.04 | 44.3 | |
| PET-DINOOpen Source=Y, Prompt Type=Visual-G, Backbone=Swin-T, VLM sup.=-, Data Size=0.6M, Training Data=O3652026.04 | 40.3 | |
| T-Rex2Open Source=N, Prompt Type=Visual-G, Backbone=Swin-T, VLM sup.=CLIP, Data Size=3.1M, Training Data=O365, OpenImages, HierText, CrowdHuman, SA-1B2026.04 | 38.8 | |
| PET-DINOOpen Source=Y, Prompt Type=Visual-G, Backbone=Swin-T, VLM sup.=-, Data Size=2.5M, Training Data=O365, OpenImages, V3Det2026.04 | 38.4 |