Semi-Automatic 3D Hand Pose Annotation

Author: Markus Oberweger oberweger@icg.tugraz.at

Requirements:

• OS:
  • Ubuntu 14.04
• via Ubuntu package manager:
  • python2.7
  • python-matplotlib
  • python-scipy
  • python-pil
  • python-numpy
  • python-vtk6
  • python-opencv2
  • python-qt4
  • python-pip
• via pip install:
  • progressbar
  • psutil
  • theano (0.8)

For a description of our method see:

M. Oberweger, G. Riegler, P. Wohlhart, and V. Lepetit. Efficiently Creating 3D Training Data for Fine Hand Pose Estimation. In Conference on Computer Vision and Pattern Recognition (CVPR), 2016.

Setup:

• Read the included manual on how to use our method
• Put dataset files into ./data
• Download SIFTFlow and put it into ./src/etc/sift_flow
• Goto ./src and see the main file main_blender_semiauto.py how to handle the API
• Goto ./src and see the main files main_labeling_pose.py and main_labeling_detect.py how to do the annotation for your own dataset