Accelerometer-based gait recognition using signature points
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
bin init version Dec 20, 2015
paper init version Dec 20, 2015
script init version Dec 20, 2015
.gitignore Initial commit Dec 20, 2015
LICENSE Initial commit Dec 20, 2015
README.md More tips Sep 21, 2016
get_data.sh init version Dec 20, 2015

README.md

Gait-SP: Accelerometer-based gait recognition using signature points

This is the code for the following paper:

Yuting Zhang, Gang Pan, Kui Jia, Minlong Lu, Yueming Wang, Zhaohui Wu, “Accelerometer-based Gait Recognition by Sparse Representation of Signature Points with Clusters”, IEEE Transactions on Cybernetics, vol. 45, no. 9, pp. 1864 – 1875, Sept. 2015.

Preparation

Compile MinMaxSelection toolbox:

$ cd bin/MinMaxSelection/MinMaxSelection
$ matlab
> minmax_install.m

You can refer to the webpage of the MinMaxSelection toolbox for more information and missing files for windows.

Compile OMP toolbox:

$ cd bin/ompbox/private
$ matlab
> make

Get extra data, including a MAT version of the ZJU-GaitAcc dataset and some cached experimental results (which can be removed when running from scratch):

$ ./get_data.sh	

Run

$ cd script
$ matlab
> main

After this, you can run the scripts in the current folder to replicate the results in our paper.

Remark

This is a beta version. Please feel free to report bugs to me.