SuperPoint 环境配置

SuperPoint  https://github.com/magicleap/SuperPointPretrainedNetwork

1、安装 anaconda

     https://www.anaconda.com/products/individual

   指令下载  wget https://repo.anaconda.com/archive/Anaconda3-2020.02-Linux-x86_64.sh   

     https://blog.csdn.net/weixin_42056625/article/details/88971708

     https://blog.csdn.net/xiaerwoailuo/article/details/70054429

       conda相关指令使用：

         https://blog.csdn.net/zhayushui/article/details/80433768?utm_medium=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1&depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromMachineLearnPai2-1

                          

2、创建一个名为pytorch的虚拟空间，并安装python 3.6

       conda create -n pytorch python=3.6

3、进(出)入虚拟空间 pytorch

     source(conda) activate(deactivate) pytorch

4、安装pytorch

     conda install pytorch torchvision cpuonly -c pytorch

     GPU上   conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0

5、安装opencv

   conda install -c menpo opencv

   如果cv2.imshow显示相关报错 ，需要再执行该命令：pip（pip3） install opencv-contrib-python

   https://blog.csdn.net/Polaris47/article/details/94560845?utm_medium=distribute.pc_relevant.none-task-blog-baidujs-5

    https://www.cnblogs.com/leoking01/p/8301724.html

    或安装opencv 3.4.1 试试