Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/busyyang/keras-yolo3-ippg
https://github.com/busyyang/keras-yolo3-ippg
Last synced: about 1 month ago
JSON representation
- Host: GitHub
- URL: https://github.com/busyyang/keras-yolo3-ippg
- Owner: busyyang
- License: mit
- Created: 2020-11-24T03:29:09.000Z (about 4 years ago)
- Default Branch: master
- Last Pushed: 2020-11-24T03:41:13.000Z (about 4 years ago)
- Last Synced: 2023-10-19T21:08:16.498Z (about 1 year ago)
- Language: Python
- Size: 3.5 MB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
数据集下载地址[百度网盘(提取码:ippg)](https://pan.baidu.com/s/1odgmskNaLG-jwwanlbsNHA),下载后解压到`wrist_images`文件夹下。
在[YOLO](https://pjreddie.com/darknet/yolo/)下载一个weights文件,放在跟目录下,命名为yolov3.weights。
将weights文件转化为模型文件:
~~~
python convert.py yolov3.cfg yolov3.weights model_data/yolo.h5
~~~
运行
~~~
python train.py
~~~
进行训练。训练完成后,运行
~~~
python yolo_inference_miou.py --image
~~~
进行模型评价。测试结果:平均IOU为0.66,最大IOU为0.95。数据集只有较少的图片,扩大数据集的数量可得到更好的结果。