Feature Space Reduction for Multimodal Human Activity Recognition
by , ,
Abstract:
This work describes the implementation, optimization, and evaluation of a Human Activity Recognition (HAR) system using 21-channel biosignals. These biosignals capture multiple modalities, such as motion and muscle activity based on two 3D-inertial sensors, one 2D-goniometer, and four electromyographic sensors. We start with an early fusion, HMM-based recognition system which discriminates 18 human activities at 91% recognition accuracy. We then optimize preprocessing with a feature space reduction and feature vector stacking. For this purpose, a Linear Discriminant Analysis (LDA) was performed based on HMM state alignments. Our experimental results show that LDA feature space reduction improves recognition accuracy by four percentage points while stacking feature vectors currently does not show any positive effects. To the best of our knowledge, this is the first work on feature space reduction in a HAR system using various biosensors integrated into a knee bandage recognizing a dive rse set of activities.
Download PDF
PDF URL: https://www.csl.uni-bremen.de/cms/images/documents/publications/HartmannLiuSchultz_Biosignals2020.pdf
Reference:
Feature Space Reduction for Multimodal Human Activity Recognition (Yale Hartmann, Hui Liu, Tanja Schultz), In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 4: BIOSIGNALS, SCITEPRESS - Science and Technology Publications, 2020.
Bibtex Entry:
@inproceedings{hartmann2020feature_space,
  title = {Feature Space Reduction for Multimodal Human Activity Recognition},
  author = {Hartmann, Yale and Liu, Hui and Schultz, Tanja},
  booktitle = {Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 4: BIOSIGNALS},
  pages = {135--140},
  organization = {INSTICC},
  publisher = {SCITEPRESS - Science and Technology Publications},
  year = {2020},
  isbn = {978-989-758-398-8},
  issn = {2184-4305},
  doi = {10.5220/0008851401350140},
  url = {https://www.csl.uni-bremen.de/cms/images/documents/publications/HartmannLiuSchultz_Biosignals2020.pdf},
  abstract = {This work describes the implementation, optimization, and evaluation of a Human Activity Recognition (HAR) system using 21-channel biosignals. These biosignals capture multiple modalities, such as motion and muscle activity based on two 3D-inertial sensors, one 2D-goniometer, and four electromyographic sensors. We start with an early fusion, HMM-based recognition system which discriminates 18 human activities at 91% recognition accuracy. We then optimize preprocessing with a feature space reduction and feature vector stacking. For this purpose, a Linear Discriminant Analysis (LDA) was performed based on HMM state alignments. Our experimental results show that LDA feature space reduction improves recognition accuracy by four percentage points while stacking feature vectors currently does not show any positive effects. To the best of our knowledge, this is the first work on feature space reduction in a HAR system using various biosensors integrated into a knee bandage recognizing a dive rse set of activities.}
}