Object Detection on VisDrone
58.3mAP50Cascaded Zoom-in
Evaluation Results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Cascaded Zoom-inVenue=CVPR'23, FPS=8.42024.01 | 58.3 | 26.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RT-DETR-lightCategory=Full-image detectors, Params (M)=32.83, GFLOPs=54.022026.04 | 58.3 | — | — | 37.8 | — | — | — | — | — | — | — | — | — | — | — | |
| B2BDetParams (M)=27.7, FLOPS (B)=109.5, FPS=27.32024.01 | 52.5 | 38.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| DFIR-DETRPara=11.7, GFLOPs=41.2, FPS=40.12025.12 | 51.6 | — | — | 31.6 | 56.6 | 39.9 | 25.1 | 83.6 | 55.4 | 57.8 | 35.6 | 36.4 | 69.6 | 55.6 | — | |
| YOLOv8nCategory=Full-image detectors, Params (M)=3.01, GFLOPs=4.102026.04 | 50.5 | — | — | 31.4 | — | — | — | — | — | — | — | — | — | — | — | |
| Generative Spatiotemporal Data AugmentationBaseline=Trad Aug, Multiplier=40x2025.12 | 49.6 | — | — | 31.3 | — | — | — | — | — | — | — | — | — | — | — | |
| Generative Spatiotemporal Data Augmentation + ControlAugBaseline=Trad Aug, Ours Multiplier=4x, ControlAug Multiplier=4x2025.12 | 49.1 | — | — | 30.8 | — | — | — | — | — | — | — | — | — | — | — | |
| TPH-YOLOv5Venue=ICCV'21, Params (M)=60, FLOPS (B)=207, FPS=18.52024.01 | 48.9 | 33.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Generative Spatiotemporal Data AugmentationBaseline=Trad Aug, Multiplier=8x2025.12 | 48.9 | — | — | 30.7 | — | — | — | — | — | — | — | — | — | — | — | |
| Generative Spatiotemporal Data Augmentation + InstaGenBaseline=Trad Aug, Ours Multiplier=4x, InstaGen Multiplier=4x2025.12 | 48.9 | — | — | 30.8 | — | — | — | — | — | — | — | — | — | — | — | |
| UAVGenDetector=RemDet-X2026.04 | 48.7 | — | — | 30.2 | — | — | — | — | — | — | — | — | — | — | 31.1 | |
| ControlAugBaseline=Trad Aug, Multiplier=8x2025.12 | 48.5 | — | — | 30.5 | — | — | — | — | — | — | — | — | — | — | — | |
| RT-DETRvariant=base, Para=19.9, GFLOPs=57.0, FPS=48.12025.12 | 48.2 | — | — | 29.1 | 53.2 | 37.3 | 25.1 | 81.3 | 53.9 | 56.8 | 31.8 | 24.7 | 67.4 | 50.1 | — | |
| Real onlyDetector=RemDet-X2026.04 | 48.1 | — | — | 29.8 | — | — | — | — | — | — | — | — | — | — | 30.8 | |
| GeodiffusionDetector=RemDet-X2026.04 | 47.8 | — | — | 29.4 | — | — | — | — | — | — | — | — | — | — | 30.2 | |
| InstaGenBaseline=Trad Aug, Multiplier=8x2025.12 | 47.7 | — | — | 29.8 | — | — | — | — | — | — | — | — | — | — | — | |
| GLIGENDetector=RemDet-X2026.04 | 47.7 | — | — | 29.4 | — | — | — | — | — | — | — | — | — | — | 30.2 | |
| SCA-YOLOVenue=VC'23, Params (M)=65.1, FLOPS (B)=142.3, FPS=24.22024.01 | 47.4 | 32.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AeroGenDetector=RemDet-X2026.04 | 46.9 | — | — | 28.8 | — | — | — | — | — | — | — | — | — | — | 29.6 | |
| Traditional AugmentationAugmentation=None/Baseline2025.12 | 45.9 | — | — | 28.5 | — | — | — | — | — | — | — | — | — | — | — | |
| YOLOv11mPara=20.1, GFLOPs=68.0, FPS=64.52025.12 | 43.4 | — | — | 26.5 | 48 | 36.2 | 18.2 | 82 | 48.6 | 40 | 32.8 | 19.3 | 60.4 | 49.3 | — | |
| YOLOv9mPara=20.0, GFLOPs=76.8, FPS=62.22025.12 | 42 | — | — | 27.6 | 17.9 | 37.1 | 17.8 | 82.1 | 50.6 | 43.6 | 37.4 | 20.8 | 62.7 | 49.8 | — | |
| YOLOv8mPara=25.9, GFLOPs=78.9, FPS=56.32025.12 | 41.9 | — | — | 25.1 | 44.7 | 34.7 | 15.9 | 80.8 | 45.9 | 39 | 31.6 | 16.2 | 63.1 | 47.5 | — | |
| YOLOv10mPara=15.4, GFLOPs=59.1, FPS=65.72025.12 | 41.5 | — | — | 25.3 | 43.6 | 35 | 15.4 | 80.5 | 48 | 40.1 | 29.9 | 18.1 | 58.3 | 46.6 | — | |
| FCOSPara=32.0, GFLOPs=203.82025.12 | 40.2 | — | — | 26.1 | 45.6 | 34.6 | 19.4 | 78.3 | 44.2 | 38.9 | 28.2 | 12.9 | 52.5 | 47.8 | — | |
| Faster R-CNNPara=33.1, GFLOPs=208.1, FPS=16.52025.12 | 39.5 | — | — | 25.6 | 39.2 | 33 | 20.8 | 74.1 | 45.8 | 39 | 30.2 | 15.1 | 55.3 | 42.6 | — | |
| Cascade R-CNNPara=69.3, GFLOPs=241.7, FPS=13.32025.12 | 39.1 | — | — | 26 | 39.1 | 31.3 | 19.3 | 74.2 | 46.3 | 39.2 | 30.8 | 13.8 | 55.1 | 42 | — | |
| CA-YOLOGFLOPs=13.50, Params M=4.04, Model Size MB=8.072026.02 | 38.4 | — | 42 | 23.06 | — | — | — | — | — | — | — | — | — | — | — | |
| MSA-YOLOVenue=Sensors'23, Params (M)=58.3, FPS=25.62024.01 | 34.7 | 21.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| YOLO11nCategory=Full-image detectors, Params (M)=2.59, GFLOPs=3.232026.04 | 34.2 | — | — | 19.8 | — | — | — | — | — | — | — | — | — | — | — | |
| YOLOv5mPara=19.9, GFLOPs=64.2, FPS=60.02025.12 | 33.8 | — | — | 18.9 | 42.6 | 33.5 | 11.5 | 74.3 | 34.8 | 30.5 | 18.4 | 10.9 | 40.8 | 40.9 | — | |
| YOLOv8nGFLOPs=8.20, Params M=3.01, Model Size MB=5.982026.02 | 33.5 | — | 38 | 19.44 | — | — | — | — | — | — | — | — | — | — | — | |
| RetinaNetPara=36.5, GFLOPs=210.0, FPS=41.32025.12 | 32.1 | — | — | 20.6 | 32.1 | 25.2 | 13 | 71.4 | 37.8 | 35.9 | 21.5 | 9.7 | 43.3 | 31.9 | — | |
| YOLOv4Venue=arXiv'20, Params (M)=64.2, FPS=22.82024.01 | 30.7 | 18.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NanoDet-PlusCategory=Full-image detectors, Params (M)=4.18, GFLOPs=1.852026.04 | 24 | — | — | 14 | — | — | — | — | — | — | — | — | — | — | — | |
| RetinaNet-R50Category=Full-image detectors, Params (M)=32.36, GFLOPs=~1452026.04 | 6.53 | — | — | 4.19 | — | — | — | — | — | — | — | — | — | — | — |