Object Detection on CAMEL
51.1APThermal-Det
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Thermal-DetZero-shot=true, Backbone=Swin-T2026.05 | 51.1 | 75.8 | 52.5 | |
| Thermal-DetSetup=Zero-shot, Backbone=Swin-T, Pre-training=Synthetic GroundingCap-1M, M3FD, MMPD, MS22026.05 | 51.1 | 75.8 | 52.5 | |
| G-DINOVenue=ECCV’24, Zero-shot=true, Backbone=Swin-T2026.05 | 48.2 | 72.9 | 54.3 | |
| G-DINOSetup=Zero-shot, Backbone=Swin-T, Pre-training=O365, GoldG, Cap4M2026.05 | 48.2 | 72.9 | 54.3 | |
| G-DINOSetup=Fine-tuning (FT), Backbone=Swin-T2026.05 | 47.8 | 68.4 | 53.3 | |
| LLMDetVenue=CVPR’25, Zero-shot=true, Backbone=Swin-T2026.05 | 38.3 | 56 | 43.9 | |
| LLMDetSetup=Zero-shot, Backbone=Swin-T, Pre-training=GroundingCap-1M2026.05 | 38.3 | 56 | 43.9 | |
| MM-GDINOZero-shot=true, Backbone=Swin-T2026.05 | 27.3 | 36.9 | 34 | |
| MM-GDINOSetup=Zero-shot, Backbone=Swin-T, Pre-training=O365, GoldG, GRIT, V3Det2026.05 | 27.3 | 36.9 | 34 | |
| T-Rex2Venue=ECCV’24, Zero-shot=true, Backbone=Swin-T2026.05 | 21.3 | 34.5 | 27.4 | |
| T-Rex2Setup=Zero-shot, Backbone=Swin-T, Pre-training=10M data from various resources2026.05 | 21.3 | 34.5 | 27.4 | |
| YOLO-WorldVenue=CVPR’24, Zero-shot=true, Backbone=Swin-T2026.05 | 19.7 | 33.6 | 25.6 | |
| YOLO-WorldSetup=Zero-shot, Backbone=YOLOv8-L, Pre-training=O365, GoldG, CC3M2026.05 | 19.7 | 33.6 | 25.6 | |
| GLIPVenue=NeurIPS’22, Zero-shot=true, Backbone=Swin-T2026.05 | 18.6 | 32.4 | 24.2 | |
| GLIPSetup=Zero-shot, Backbone=Swin-T, Pre-training=O365, GoldG, Cap4M2026.05 | 18.6 | 32.4 | 24.2 |