7th International Conference on Pervasive Computing Technologies for Healthcare

Research Article

Towards Robust Activity Recognition for Everyday Life: Methods and Evaluation

Download13 downloads
  • @INPROCEEDINGS{10.4108/icst.pervasivehealth.2013.251928,
        author={Attila Reiss and Gustaf Hendeby and Didier Stricker},
        title={Towards Robust Activity Recognition for Everyday Life: Methods and Evaluation},
        proceedings={7th International Conference on Pervasive Computing Technologies for Healthcare},
        publisher={IEEE},
        proceedings_a={PERVASIVEHEALTH},
        year={2013},
        month={5},
        keywords={activity recognition evaluation robustness},
        doi={10.4108/icst.pervasivehealth.2013.251928}
    }
    
  • Attila Reiss
    Gustaf Hendeby
    Didier Stricker
    Year: 2013
    Towards Robust Activity Recognition for Everyday Life: Methods and Evaluation
    PERVASIVEHEALTH
    ICST
    DOI: 10.4108/icst.pervasivehealth.2013.251928
Attila Reiss1,*, Gustaf Hendeby2, Didier Stricker1
  • 1: German Research Center for Artificial Intelligence (DFKI)
  • 2: Linköping University
*Contact email: attila.reiss@dfki.de

Abstract

The monitoring of physical activities under realistic, everyday life conditions thus while an individual follows his regular daily routine is usually neglected or even completely ignored. Therefore, this paper investigates the development and evaluation of robust methods for everyday life scenarios, with focus on the task of aerobic activity recognition. Two important aspects of robustness are investigated: dealing with various (unknown) other activities and subject independency. Methods to handle these issues are proposed and compared, a thorough evaluation simulates usual everyday scenarios of the usage of activity recognition applications. Moreover, a new evaluation technique is introduced (leave-one-other-activity-out) to simulate when an activity recognition system is used while performing a previously unknown activity. Through applying the proposed methods it is possible to design a robust physical activity recognition system with the desired generalization characteristic.