Skip to content

Latest commit

 

History

History
48 lines (35 loc) · 1.43 KB

README.md

File metadata and controls

48 lines (35 loc) · 1.43 KB
Raw

SETUP for OPENPOSE and 3D-POSE-BASELINE on Ubuntu 16.04 / 180505

Note: OpenPose alone can record 3D coordinates with a multi-camera setup. However the 3d-pose-baseline method allows generating 3D coordinates for multiple subjects from a single camera.

  1. Clone OpenPose:
    https://github.com/CMU-Perceptual-Computing-Lab/openpose

  2. Install CUDA, CUDNN, and OpenCV

./ubuntu/install_cuda.sh
./ubuntu/install.cudnn.sh
sudo apt-get install libopencv-dev python-opencv
  1. Download models
./models/getModels.sh
  1. Install Caffe and build OpenPose
./ubuntu/install_caffe_and_openpose_if_cuda8.sh

  1. Sample OpenPose shell scripts for Ubuntu and Windows.

  2. Clone 3D-pose-baseline (fork with OpenPose converter script):
    https://github.com/ArashHosseini/3d-pose-baseline

  3. Install Python dependencies:

pip install h5py
pip install tensorflow
pip install imageio
pip install matplotlib
  1. Download models:

(Place in ./experiments/)
https://drive.google.com/file/d/0BxWzojlLp259MF9qSFpiVjl0cU0/view?usp=sharing

(Place in ./data/h36m/)
https://www.dropbox.com/s/e35qv3n6zlkouki/h36m.zip

  1. Create a directory ./png for output.

  2. Run a sample shell script.