To utilise OpenPose library for hand movement trajectory tracking
- Firstly please intall the tf-openpose model following the instructions of the link: https://github.com/ildoonet/tf-pose-estimation. Thanks to the amazing work of Ildoo Kim, that translated most code of the OpenPose Library to Python.
- Then download the code for hand movenment tracking of this repository. Hopefully you will get the results as shown below. 😉
For the reference, this model has been developed and tested in the the CPU desktop of 8 GB RAM 3.00 GHz Intel Core i5-4590SCPU processor, also on a GPU desktop with two NVIDIA GeForce GTX 1080Ti adapter cards and 3.3 GHz In-tel Core i9-7900X CPU with 16 GB RAM.
for CPU environment the model was implemeted in:
- Tensorflow 1.11
- python 3.6.5
- OpenCV 3.3.1
for GPU environment the model was implemeted in
- Tensorflow 1.12
- Python 3.6.8
- OpenCV 3.4.2
@inproceedings{liang2019handtracking,
author = {X. Liang, E. Kapetanios, B. Woll and A. Angelopoulou},
booktitle = {Cross Domain Conference for Machine
Learning and Knowledge Extraction (CD-MAKE2019)},
title = {Real Time Hand Movement Trajectory Tracking for Enhancing
Dementia Screening in Ageing Deaf Signers of British Sign Language},
year = {2019}
}