Jingyuan Cheng, Bo Zhou, Mathias Sundholm, Paul Lukowicz


In this paper we present a very simple chair system for activity recognition. With just 4 pressure sensors under the chair legs, the system provides sufficient information not only for detecting body postures (0.839 accuracy for 5 subjects and 7 classes) but also for minor activities like typing the keyboard (0.853 accuracy for 5 subjects and 5 classes). Beyond single small activities, high-level daily activities like working in front of computers, having meals can be reasonably distinguished from each other. In a dataset of 105.6 hours recorded from 4 subjects, we achieved 0.761 accuracy for 7 classes.


@inproceedings {Cheng:Smart:2013:7002,
	number = {}, 
	month = {}, 
	year = {2013}, 
	title = {Smart Chair: What can simple pressure sensors under the chairs legs tell us about user activity?}, 
	journal = {}, 
	volume = {}, 
	pages = {}, 
	publisher = {IARIA XPS Press}, 
	author = {Jingyuan Cheng, Bo Zhou, Mathias Sundholm, Paul Lukowicz}, 
	keywords = {}