Research Article
Metric Based Automatic Event Segmentation
@INPROCEEDINGS{10.1007/978-3-642-36632-1_8, author={Yuwen Zhuang and Mikhail Belkin and Simon Dennis}, title={Metric Based Automatic Event Segmentation}, proceedings={Mobile Computing, Applications, and Services. 4th International Conference, MobiCASE 2012, Seattle, WA, USA, October 11-12, 2012. Revised Selected Papers}, proceedings_a={MOBICASE}, year={2013}, month={2}, keywords={Triaxial accelerometer Event segmentation Lifelog data}, doi={10.1007/978-3-642-36632-1_8} }- Yuwen Zhuang
Mikhail Belkin
Simon Dennis
Year: 2013
Metric Based Automatic Event Segmentation
MOBICASE
Springer
DOI: 10.1007/978-3-642-36632-1_8
Abstract
This paper describes a metric-based model for event segmentation of sensor data recorded by a mobile phone worn around subjects’ necks during their daily life. More specifically, we aim at detecting human daily event boundaries by analysing the recorded triaxial accelerometer signals and images sequence (lifelog data). In the experiments, different signal representations and three boundary detection models are evaluated on a corpus of 2 subjects over total 24 days. The contribution of this paper is three-fold. First, we find that using accelerometer signals can provide much more reliable and significantly better performance than using image signals with MPEG-7 low level features. Second, the models using the accelerometer data based on the world’s coordinates system can provide equally or even much better performance than using the accelerometer data based on the device’s coordinates system. Finally, our proposed model has a better performance than the state of the art system [1].