Distributed segmentation and classification of human actions using a wearable motion sensor network
Authors:
- Philip KuryloskiPKPhilip Kuryloski
- Roozbeh JafariRJRoozbeh Jafari
- 1 June 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21607508,p. 1-8
Abstract
We propose a distributed recognition method to classify human actions using a low-bandwidth wearable motion sensor network. Given a set of pre-segmented motion sequences as training examples, the algorithm simultaneously segments and classifies human actions, and it also rejects outlying actions that are not in the training set. The classification is distributedly operated on individual sensor nodes and a base station computer. We show that the distribution of multiple action classes satisfies a mixture subspace model, one sub-space for each action class. Given a new test sample, we seek the sparsest linear representation of the sample w.r.t. all training examples. We show that the dominant coefficients in the representation only correspond to the action class of the test sample, and hence its membership is encoded in the representation. We further provide fast linear solvers to compute such representation via l1-minimization. Using up to eight body sensors, the algorithm achieves state-of-the-art 98.8% accuracy on a set of 12 action categories. We further demonstrate that the recognition precision only decreases gracefully using smaller subsets of sensors, which validates the robustness of the distributed framework.Keywords
This publication has 12 references indexed in Scilit:
- Groggy Wakeup - Automated Generation of Power-Efficient Detection Hierarchies for Wearable Sensors2007
- A compact, high-speed, wearable sensor network for biomotion capture and interactive media2007
- Data compression algorithms for energy-constrained devices in delay tolerant networks2006
- On the stability of the basis pursuit in the presence of noiseSignal Processing, 2006
- Analyzing features for activity recognition2005
- Ambulatory system for human motion analysis using a kinematic sensor: monitoring of daily physical activity in the elderlyIEEE Transactions on Biomedical Engineering, 2003
- Multi-sensor Activity Context Detection for Wearable Computing2003
- Recognizing human motion with multiple acceleration sensors2002
- Distributed recognition of patterns in time series dataCommunications of the ACM, 1998
- Validity and reliability of pedometers in habitual activity researchEuropean Journal of Applied Physiology, 1977