Dealing with sensor displacement in motion-based onbody activity recognition systems

Kai Steven Kunze, Paul Lukowicz

Research output: Chapter in Book/Report/Conference proceedingConference contribution

101 Citations (Scopus)

Abstract

We present a set of heuristics that significantly increase the robustness of motion sensor-based activity recognition with respect to sensor displacement. In this paper placement refers to the position within a single body part (e.g, lower arm). We show how, within certain limits and with modest quality degradation, motion sensorbased activity recognition can be implemented in a displacement tolerant way. We first describe the physical principles that lead to our heuristic. We then evaluate them first on a set of synthetic lower arm motions which are well suited to illustrate the strengths and limits of our approach, then on an extended modes of locomotion problem (sensors on the upper leg) and finally on a set of exercises performed on various gym machines (sensors placed on the lower arm). In this example our heuristic raises the displaced recognition rate from 24% for a displaced accelerometer, which had 96% recognition when not displaced, to 82%.

Original languageEnglish
Title of host publicationUbiComp 2008 - Proceedings of the 10th International Conference on Ubiquitous Computing
Pages20-29
Number of pages10
DOIs
Publication statusPublished - 2008
Externally publishedYes
Event10th International Conference on Ubiquitous Computing, UbiComp 2008 - Seoul, Korea, Republic of
Duration: 2008 Sep 212008 Sep 24

Other

Other10th International Conference on Ubiquitous Computing, UbiComp 2008
CountryKorea, Republic of
CitySeoul
Period08/9/2108/9/24

Fingerprint

Sensors
Accelerometers
Degradation

Keywords

  • Fitness exercises
  • Motion sensors
  • Opportunistic activity recognition
  • Sensor displacement

ASJC Scopus subject areas

  • Computer Science Applications
  • Human-Computer Interaction
  • Software

Cite this

Kunze, K. S., & Lukowicz, P. (2008). Dealing with sensor displacement in motion-based onbody activity recognition systems. In UbiComp 2008 - Proceedings of the 10th International Conference on Ubiquitous Computing (pp. 20-29) https://doi.org/10.1145/1409635.1409639

Dealing with sensor displacement in motion-based onbody activity recognition systems. / Kunze, Kai Steven; Lukowicz, Paul.

UbiComp 2008 - Proceedings of the 10th International Conference on Ubiquitous Computing. 2008. p. 20-29.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kunze, KS & Lukowicz, P 2008, Dealing with sensor displacement in motion-based onbody activity recognition systems. in UbiComp 2008 - Proceedings of the 10th International Conference on Ubiquitous Computing. pp. 20-29, 10th International Conference on Ubiquitous Computing, UbiComp 2008, Seoul, Korea, Republic of, 08/9/21. https://doi.org/10.1145/1409635.1409639
Kunze KS, Lukowicz P. Dealing with sensor displacement in motion-based onbody activity recognition systems. In UbiComp 2008 - Proceedings of the 10th International Conference on Ubiquitous Computing. 2008. p. 20-29 https://doi.org/10.1145/1409635.1409639
Kunze, Kai Steven ; Lukowicz, Paul. / Dealing with sensor displacement in motion-based onbody activity recognition systems. UbiComp 2008 - Proceedings of the 10th International Conference on Ubiquitous Computing. 2008. pp. 20-29
@inproceedings{092b9af0dc664639bdfd6393928519ce,
title = "Dealing with sensor displacement in motion-based onbody activity recognition systems",
abstract = "We present a set of heuristics that significantly increase the robustness of motion sensor-based activity recognition with respect to sensor displacement. In this paper placement refers to the position within a single body part (e.g, lower arm). We show how, within certain limits and with modest quality degradation, motion sensorbased activity recognition can be implemented in a displacement tolerant way. We first describe the physical principles that lead to our heuristic. We then evaluate them first on a set of synthetic lower arm motions which are well suited to illustrate the strengths and limits of our approach, then on an extended modes of locomotion problem (sensors on the upper leg) and finally on a set of exercises performed on various gym machines (sensors placed on the lower arm). In this example our heuristic raises the displaced recognition rate from 24{\%} for a displaced accelerometer, which had 96{\%} recognition when not displaced, to 82{\%}.",
keywords = "Fitness exercises, Motion sensors, Opportunistic activity recognition, Sensor displacement",
author = "Kunze, {Kai Steven} and Paul Lukowicz",
year = "2008",
doi = "10.1145/1409635.1409639",
language = "English",
isbn = "9781605581361",
pages = "20--29",
booktitle = "UbiComp 2008 - Proceedings of the 10th International Conference on Ubiquitous Computing",

}

TY - GEN

T1 - Dealing with sensor displacement in motion-based onbody activity recognition systems

AU - Kunze, Kai Steven

AU - Lukowicz, Paul

PY - 2008

Y1 - 2008

N2 - We present a set of heuristics that significantly increase the robustness of motion sensor-based activity recognition with respect to sensor displacement. In this paper placement refers to the position within a single body part (e.g, lower arm). We show how, within certain limits and with modest quality degradation, motion sensorbased activity recognition can be implemented in a displacement tolerant way. We first describe the physical principles that lead to our heuristic. We then evaluate them first on a set of synthetic lower arm motions which are well suited to illustrate the strengths and limits of our approach, then on an extended modes of locomotion problem (sensors on the upper leg) and finally on a set of exercises performed on various gym machines (sensors placed on the lower arm). In this example our heuristic raises the displaced recognition rate from 24% for a displaced accelerometer, which had 96% recognition when not displaced, to 82%.

AB - We present a set of heuristics that significantly increase the robustness of motion sensor-based activity recognition with respect to sensor displacement. In this paper placement refers to the position within a single body part (e.g, lower arm). We show how, within certain limits and with modest quality degradation, motion sensorbased activity recognition can be implemented in a displacement tolerant way. We first describe the physical principles that lead to our heuristic. We then evaluate them first on a set of synthetic lower arm motions which are well suited to illustrate the strengths and limits of our approach, then on an extended modes of locomotion problem (sensors on the upper leg) and finally on a set of exercises performed on various gym machines (sensors placed on the lower arm). In this example our heuristic raises the displaced recognition rate from 24% for a displaced accelerometer, which had 96% recognition when not displaced, to 82%.

KW - Fitness exercises

KW - Motion sensors

KW - Opportunistic activity recognition

KW - Sensor displacement

UR - http://www.scopus.com/inward/record.url?scp=59249086539&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=59249086539&partnerID=8YFLogxK

U2 - 10.1145/1409635.1409639

DO - 10.1145/1409635.1409639

M3 - Conference contribution

AN - SCOPUS:59249086539

SN - 9781605581361

SP - 20

EP - 29

BT - UbiComp 2008 - Proceedings of the 10th International Conference on Ubiquitous Computing

ER -