mmpose 安装及使用pretrained model 运行inferencer_demo.py Running RTMPose,ViTPose,YoloPose模型

Installation官网教程 ：

Inference with existing models — MMPose 1.3.2 documentation

Installation — MMPose 1.3.2 documentation

env:Ubuntu22.04 CUDA11.6, NVIDIA RTX3090Ti， 其他软件版本见下图：

conda create --name openmmlab python=3.8 -y
conda activate openmmlab


conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.6 -c pytorch -c nvidia

pip install -U openmim
mim install mmengine
mim install mmcv==2.1.0

mim install mmdet==3.2.0

Build MMPose from source

git clone https://github.com/open-mmlab/mmpose.git
cd mmpose
pip install -r requirements.txt
pip install -v -e .
# "-v" means verbose, or more output
# "-e" means installing a project in editable mode,
# thus any local modifications made to the code will take effect without reinstallation.

Terminal : mmpose 项目根目录下运行 RTMPose,ViTPose,YoloPose 模型

python demo/inferencer_demo.py /video_or_images_path \
   --pose2d vitpose-h   \
   --pred-out-dir ../output_dir/ \
   --skeleton-style openpose \ 
   --device cuda:0  //device 指定GPU 索引0

更多参数设置详细阅读mmpose项目代码：mmpose/demo/inferencer_demo.py 

 Available 2D model aliases and their corresponding configuration names:

python demo/inferencer_demo.py --show-alias