实战OpenPose项目2：开发环境配置与demo运行

上一篇文章：
实战OpenPose项目：开篇使用文档
ubuntu18或者使用docker
确保你的电脑环境支持NVIDIA显卡驱动。

在 Ubuntu 上使用 cudnn 7.5 的安装指南 cuda-10-0（用于 Tensorflow/Pytorch）

Remove previous versions of CUDA:

如果这一步已经安装好了cuda以及对应的NVIDIA版本就无需操作。

sudo apt-get purge nvidia*
sudo apt-get autoremove
sudo apt-get autoclean
sudo rm -rf /usr/local/cuda*

Install cuda

Go to Nvidia site here https://developer.nvidia.com/cuda-10.0-download-archive
Choose your:

1. Platform
2. Architecture
3. Distribution
4. Version
5. In the last step choose dev (network)

Download the file

# for Ubuntu18.04
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/cuda-repo-ubuntu1804_10.0.130-1_amd64.deb

Then

sudo dpkg -i cuda-repo-ubuntu1804_10.0.130-1_amd64.deb
sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/7fa2af80.pub
sudo apt-get update
sudo apt-get install cuda-10-0

Reboot

sudo reboot

Add cuda to the $path

echo 'export PATH=/usr/local/cuda-10.0/bin${PATH:+:${PATH}}' >> ~/.bashrc
echo 'export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}' >> ~/.bashrc
source ~/.bashrc
sudo ldconfig

Install cudNN

Go to the site https://developer.nvidia.com/rdp/cudnn-archive
Download cuDNN v7.5.0 (Feb 21, 2019), for CUDA 10.0, cuDNN Library for Linux. So you should have cudnn-10.0-linux-x64-v7.5.0.56.tgz
Then copy necessary files

tar -xf cudnn-10.0-linux-x64-v7.5.0.56.tgz
sudo cp -R cuda/include/* /usr/local/cuda-10.0/include
sudo cp -R cuda/lib64/* /usr/local/cuda-10.0/lib64

最后还要安装C++版本的opencv，此外需要安装cmake进行编译（终端在openpos下面直接cmake…即可进行编译，也可以打开cmake-gui进行编译，编译完成记得生成。）

demo运行

上一篇文章已经说的很清楚，这里只是补充说明环境配置 的问题。

参考文献：
https://gist.github.com/Inkognita/961005403e7c824addd225fc20377db2