CUDA11.1安装AlphaPose

依然☞范佩西 

已于 2022-12-11 13:33:50 修改

阅读量1.3k

点赞数 1

文章标签： pytorch 人工智能 python

于 2021-12-16 15:15:40 首次发布

本文链接：https://blog.csdn.net/weixin_48701154/article/details/121938154

版权

我的电脑是：ubuntu20.04 RTX3070 cuda11.1

官方安装：AlphaPose/INSTALL.md at master · MVIG-SJTU/AlphaPose · GitHub

GitHub - MVIG-SJTU/AlphaPose: Real-Time and Accurate Full-Body Multi-Person Pose Estimation&Tracking System

# 1. Create a conda virtual environment.
conda create -n alphapose python=3.7 -y
conda activate alphapose

# 2. Install PyTorch
conda install pytorch==1.8.0 torchvision==0.9.0 torchaudio==0.8.0 cudatoolkit=11.1 -c pytorch -c conda-forge

# 3. Get AlphaPose
git clone https://github.com/MVIG-SJTU/AlphaPose.git

用pycharm打开clone的AlphaPose,在终端执行
git pull origin pull/592/head 

忘了截图，中间默认点击确定就好了（这个图片怎么一直传不上来！？）




cd AlphaPose

# 4. install
export PATH=/usr/local/cuda/bin/:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64/:$LD_LIBRARY_PATH
python -m pip install cython
sudo apt-get install libyaml-dev

python setup.py build develop

出现错误

多重复几次后：

终于装好了，嘤嘤嘤。

测试图片

目标检测的模型：yolov3-spp.weights

把yolov3-spp.weights放到detector/yolo/data中

下载合适的模型：fast_res50_256x192.pth

把下载好的模型放到AlphaPose/pretrained_models文件夹下

测试命令：

python scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml --checkpoint pretrained_models/fast_res50_256x192.pth --indir /home/miao/study/3Drecon/test/alphapose/in/ --outdir /home/miao/study/3Drecon/test/alphapose/out/ --save_img