Paper Info
Title
Towards robust activity recognition for everyday life: methods and evaluation
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.
Year
DOI
Venue
2013
10.4108/icst.pervasivehealth.2013.251928
PervasiveHealth
Keywords
Field
DocType
everyday life scenario,new evaluation technique,unknown activity,activity recognition system,robust physical activity recognition,aerobic activity recognition,towards robust activity recognition,thorough evaluation,everyday life condition,physical activity,activity recognition application,information systems,patient monitoring,signal processing
Information system,Everyday life,Activity recognition,Computer science,Remote patient monitoring,Robustness (computer science),Artificial intelligence,Regular daily routine,Unknown activity,Machine learning
Conference
ISSN
Citations 
PageRank 
2153-1633
15
0.63
References 
Authors
14
3
Name
Order
Citations
PageRank
Attila Reiss141024.01
Didier Stricker21266138.03
Gustaf Hendeby321621.37