probabilistic First-Take-All feature for human activity recognition from sensor signals. This is the matlab source code for the pFTA feature and the probabilistic Temporal Order Encoding Algorithm proposed in the paper:
J. Ye, G. Qi, N. Zhuang, H. Hu and K. A. Hua, "Learning Compact Features for Human Activity Recognition via Probabilistic First-Take-All," in IEEE Transactions on Pattern Analysis and Machine Intelligence.doi: 10.1109/TPAMI.2018.2874455 URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8485314&isnumber=4359286
- A preprocessed Smpartphone-based Human Activity Dataset is provided to repeat the experiment presented in the paper.The original Dataset can be found at https://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones In the experiment, the 30 subjects of the dataset are evenly split into 5 groups and the subject-based Leave-one-out cross-validation is employed for the performance evaluation. The final result is the average of the results of the above 5 runs. Subject grouping policy is as follows,
| Groups | Subjects |
|---|---|
| Group1 | 1 2 3 4 5 6 |
| Group2 | 7 8 9 10 11 12 |
| Group3 | 13 14 15 16 17 18 |
| Group4 | 19 20 21 22 23 24 |
| Group5 | 25 26 27 28 29 30 |
-
Run the scriptRepeatExperimentResult.m to get pFTA classification results based on the optimized projections.
-
Run the scriptAutoTestRand.m to get the pFTA classification results based on random projection.
-
To repeat the results on the random projection, please set the rand seed as "default".
-
To train your own projections, run ScriptAutoTrain.m
-
The mtimesx tool is needed to run the algorithm to compute the multiplication of ND Array in matlab. I have enclosed this package in the source.
Copyright (c) Jun Ye. 2016. This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.