yolo9000论文笔记
1.19.7 mAP imagenet detection
2. 67 FPS 76.8 mAP VOC 2007
3. 40 FPS 78.6 mAP
4.物体识别目的： 更快，更准，识别更多的物体
5.数据集： Our method uses a hierarchical view of object classification that allows us to
combine distinct datasets together
6.训练方法： propose a joint training algorithm that allows us to train object detectors on both detection and classification
data
7.YOLO detection system -> YOLOv2 -> YOLO9000
2. 67 FPS 76.8 mAP VOC 2007
3. 40 FPS 78.6 mAP
4.物体识别目的： 更快，更准，识别更多的物体
5.数据集： Our method uses a hierarchical view of object classification that allows us to
combine distinct datasets together
6.训练方法： propose a joint training algorithm that allows us to train object detectors on both detection and classification
data
7.YOLO detection system -> YOLOv2 -> YOLO9000
