Purpose

This is the code the challenge"CHALEARN Gesture Challenge“. https://www.kaggle.com/c/GestureChallenge

Gist:Extended MHI + MSE

by Di WU: stevenwudi@gmail.com, 2012/03/27

Citation

If you use this toolbox as part of a research project, please cite the corresponding paper

@inproceedings{wu2012one,
  title={One Shot Learning Gesture Recognition from RGBD Images},
  author={Wu, Di and Zhu, Fan and Shao, Ling},
  booktitle={Computer Vision and Pattern Recognition Workshops (CVPRW), 2012 IEEE Computer Society Conference on},
  year={20123}
}

Dependency:

(1) mmread folder: this folder is in the original sample code to read video files which can be downloaded at: http://www.kaggle.com/c/GestureChallenge/Data (may not be necessary for newer version of matlab

Train

run_this.m is the m-file to run, simply change data_dir and resu_file to the desirable directories

Usage Instructions

To change the development or validation batches, change the line 32 &33 in prepare_final_resu.m.

Note

whether to use the lossi-compressed data or the quasi-lossless compressed data We used the quasi-lossless compressed data downloaded from the kaggle website.

Running Time