所属分类：
移动互联网开发
开发工具：Python
文件大小：8479KB
下载次数：7
上传日期：2019-02-18 10:05:55
说明： Yolo算法采用一个单独的CNN模型实现end-to-end的目标检测，整个系统如图5所示：首先将输入图片resize到448x448，然后送入CNN网络，最后处理网络预测结果得到检测的目标。相比R-CNN算法，其是一个统一的框架，其速度更快，而且Yolo的训练过程也是end-to-end的
(Yolo algorithm uses a single CNN model to realize end-to-end target detection. The whole system is shown in Fig. 5. Firstly, the input picture resize to 448x448 is sent to CNN network. Finally, the detection target is obtained by processing the network prediction results. Compared with R-CNN algorithm, it is a unified framework with faster speed, and Yolo's training process is end-to-end.)
文件列表：
YOLO_tensorflow-master, 0 , 2017-12-06
YOLO_tensorflow-master\LICENSE, 641 , 2017-12-06
YOLO_tensorflow-master\README.md, 2532 , 2017-12-06
YOLO_tensorflow-master\YOLO_face_tf.py, 10222 , 2017-12-06
YOLO_tensorflow-master\YOLO_small_tf.py, 10628 , 2017-12-06
YOLO_tensorflow-master\YOLO_tiny_tf.py, 9863 , 2017-12-06
YOLO_tensorflow-master\YOLO_weight_extractor, 0 , 2017-12-06
YOLO_tensorflow-master\YOLO_weight_extractor\Readme.md, 864 , 2017-12-06
YOLO_tensorflow-master\YOLO_weight_extractor\YOLO_weight_extractor.tar.gz, 8558420 , 2017-12-06
YOLO_tensorflow-master\test, 0 , 2017-12-06
YOLO_tensorflow-master\test\person.jpg, 113880 , 2017-12-06
YOLO_tensorflow-master\weights, 0 , 2017-12-06
YOLO_tensorflow-master\weights\put_weight_file_here.txt, 0 , 2017-12-06
