Segnet分割网络caffe教程（一）：A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation 视频语义分割

╰+攻爆jí腚メ 2022-02-28 05:20 250阅读 0赞

# 前提： #

已经安装好caffe-segnet，并且下载了SegNet-Tutorial-master。

如没有，请参考安装教程：

Caffe-SegNet安装笔记（基于Python2.7）：[https://blog.csdn.net/zhuiqiuzhuoyue583/article/details/87256632][https_blog.csdn.net_zhuiqiuzhuoyue583_article_details_87256632]

注意电脑的显存必须是2G，或以上，否则总是显示内存溢出，我的笔记本就是1G的显卡，运行失败，在台式机上才运行成功。

下载模型：segnet\_weights\_driving\_webdemo.caffemodel，下载后，将其放在“SegNet/Example\_Model”文件夹中。

模型下载地址：

[http://mi.eng.cam.ac.uk/~agk34/resources/SegNet/][http_mi.eng.cam.ac.uk_agk34_resources_SegNet]

或者：https://github.com/alexgkendall/SegNet-Tutorial/blob/master/Example\_Models/segnet\_model\_zoo.md

1.进入到SegNet(原名：SegNet-Tutorial-master)文件的Script文件夹中，修改webcam\_demo.py  
第十四行：

#caffe_root = '/SegNet/caffe-segnet/'
    caffe_root = '/home/zqzy/SegNet /caffe-segnet/'
    #cap = cv2.VideoCapture(0) # Change this to your webcam ID, or file name for your video file
    cap = cv2.VideoCapture("/home/zqzy/SegNet /Scripts/0006R0.MXF")

这里的0006R0.XMF是一段视频，将其格式修改为XFM即可。

2.运行命令：进入到SegNet文件夹

python Scripts/webcam_demo.py --model Example_Models/segnet_model_driving_webdemo.prototxt --weights Example_Models/segnet_weights_driving_webdemo.caffemodel --colours Scripts/camvid12.png

![loading.png][]

说明： 运行命令中的caffemodel模型可以替换的，有室内的训练模型，也有室外的训练模型。

模型下载地址：

[http://mi.eng.cam.ac.uk/~agk34/resources/SegNet/][http_mi.eng.cam.ac.uk_agk34_resources_SegNet]

或者：https://github.com/alexgkendall/SegNet-Tutorial/blob/master/Example\_Models/segnet\_model\_zoo.md

相关网站： https://github.com/alexgkendall/SegNet-Tutorial

http://mi.eng.cam.ac.uk/projects/segnet/\#code

http://blog.csdn.net/u014451076/article/details/70741629?locationNum=1&fps=1

https://askubuntu.com/questions/784392/issues-with-nvidia-graphics-driver-and-cuda-after-apt-get-upgrade 搞定~~~

参考：

[http://www.voidcn.com/article/p-xsmwtymr-bqh.html][http_www.voidcn.com_article_p-xsmwtymr-bqh.html]

[https://blog.csdn.net/caicai2526/article/details/77169871][https_blog.csdn.net_caicai2526_article_details_77169871]

[https_blog.csdn.net_zhuiqiuzhuoyue583_article_details_87256632]: https://blog.csdn.net/zhuiqiuzhuoyue583/article/details/87256632
[http_mi.eng.cam.ac.uk_agk34_resources_SegNet]: http://mi.eng.cam.ac.uk/~agk34/resources/SegNet/
[loading.png]: http://img.voidcn.com/vcimg/static/loading.png
[http_www.voidcn.com_article_p-xsmwtymr-bqh.html]: http://www.voidcn.com/article/p-xsmwtymr-bqh.html
[https_blog.csdn.net_caicai2526_article_details_77169871]: https://blog.csdn.net/caicai2526/article/details/77169871