Holger Junker, Oliver Amft, Paul Lukowicz, Gerhard Tröster


We present a method for spotting sporadically occurring gestures in a continuous data stream from body-worn inertial sensors. Our method is based on a natural partitioning of continuous sensor signals and uses a two-stage approach for the spotting task. In a first stage, signal sections likely to contain specific motion events are preselected using a simple similarity search. Those preselected sections are then further classified in a second stage, exploiting the recognition capabilities of hidden Markov models. Based on two case studies, we discuss implementation details of our approach and show that it is a feasible strategy for the spotting of various types of motion events.   [Download]


@article {Junker:Gesture:2008:6518,
	number = {6}, 
	month = {}, 
	year = {2008}, 
	title = {Gesture spotting with body-worn inertial sensors to detect user activities}, 
	journal = {Pattern Recognition (PR)}, 
	volume = {41}, 
	pages = {2010-2024}, 
	publisher = {Elsevier Science Inc., New York, NY, USA}, 
	author = {Holger Junker, Oliver Amft, Paul Lukowicz, Gerhard Tröster}, 
	keywords = {Activity recognition, Automatic dietary monitoring, Event detection, Gesture spotting, Natural gesture segmentation}