Consistent-Teacher: Towards Reducing Inconsistent Pseudo-targets in Semi-supervised Object Detection
mAP: 48.20
Papers with Code · CC BY-SA 4.0
Semi-Supervised Object Detection
Semi-supervised object detection uses both labeled data and unlabeled data for training. It not only reduces the annotation burden for training high-performance object detectors but also further improves the object detector by using a large number of unlabeled data.
논문
mAP: 46.2
mAP: 55.2
mAP: 36.95detector: FCOS-Res50
mAP: 46.1
mAP: 44.0
mAP: 50.5
mAP: 42.79
mAP: 39.2
mAP: 34.58
mAP: 38.5detector: FCOS-Res50
mAP: 37.9detector: YOLOv5-L
mAP: 44.9
mAP: 40.2
mAP: 35.30±0.15detector: FasterRCNN-Res50
mAP: 42.11
mAP: 34.1
mAP: 43.3
mAP: 34.82detector: FasterRCNN-Res50
10%: 44.3
mAP: 41.3
mAP: 35.08±0.02detector: FCOS-Res50