Fusion of String-Matched Templates for Continuous Activity Recognition

Authors:
Thomas Stiefmeier;Daniel Roggen;Gerhard Troster
Affiliations:
Wearable Computing Lab, ETH Zürich, Switzerland, stiefmeier@ife.ee.ethz.ch;Wearable Computing Lab, ETH Zürich, Switzerland, droggen@ife.ee.ethz.ch;Wearable Computing Lab, ETH Zürich, Switzerland, troester@ife.ee.ethz.ch
Venue:
ISWC '07 Proceedings of the 2007 11th IEEE International Symposium on Wearable Computers
Year:
2007

Citing 0
Cited 6

Network-adaptive management of computation energy in wireless sensor networks

Proceedings of the 2010 ACM Symposium on Applied Computing
Activity recognition from on-body sensors: accuracy-power trade-off by dynamic sensor selection

EWSN'08 Proceedings of the 5th European conference on Wireless sensor networks
A bag-of-features-based framework for human activity representation and recognition

Proceedings of the 2011 international workshop on Situation activity & goal awareness
Motion primitive-based human activity recognition using a bag-of-features approach

Proceedings of the 2nd ACM SIGHIT International Health Informatics Symposium
Network-Level Power-Performance Trade-Off in Wearable Activity Recognition: A Dynamic Sensor Selection Approach

ACM Transactions on Embedded Computing Systems (TECS)
Detecting leisure activities with dense motif discovery

Proceedings of the 2012 ACM Conference on Ubiquitous Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a new method for continuous activity recognition based on fusion of string-matched activity templates. The underlying segmentation and spotting approach is carried out on several symbol streams in parallel. These streams represent motion trajectories of body limbs in Cartesian space, acquired from body-worn inertial sensors. First results of our method in a highly complex real-world application are presented. 8 subjects performed 3800 activity instances of a checking procedure in car assembly adding up to 480 minutes of recordings. Selecting 6 activity classes with 468 occurrences for first investigations, we obtained an accuracy of up to 87% for the user-dependent case.