Activity recognition for Smart City scenarios: Google Play Services vs. MoST facilities

Author

Cardone, Giuseppe ; Cirri, Andrea ; Corradi, Antonio ; Foschini, Luca ; Montanari, Rebecca

Author_Institution

Dipt. di Inf. - Scienza e Ing., Univ. of Bologna, Bologna, Italy

fYear

2014

fDate

23-26 June 2014

Firstpage

1

Lastpage

6

Abstract

The ever increasing diffusion of smartphones today equipped with several physical and virtual sensors allow to directly collect information about surrounding physical and logical context that range from monitoring current social pulse of individuals and entire communities to detecting user current physical activity. Enabling those advanced sensing capabilities requires complex signal processing, machine learning, and resource management algorithms that are often beyond the skills of many mobile app developers. This paper describes the relevance of these facilities for mobile crowdsensing applications in Smart City scenarios and presents our solution for activity detection, comparing it with the reference implementations provided by Google as part of the Google Play Services library.

Keywords

image recognition; learning (artificial intelligence); mobile computing; object detection; search engines; Google play services library; MoST facilities; activity recognition; advanced sensing capability; complex signal processing; logical context; machine learning; mobile app developers; mobile crowdsensing; physical context; resource management algorithms; smart city scenarios; smart phones; social pulse monitoring; user current physical activity detection; virtual sensors; Accelerometers; Google; Legged locomotion; Libraries; Pipelines; Sensors; Smart phones;

fLanguage

English

Publisher

ieee

Conference_Titel

Computers and Communication (ISCC), 2014 IEEE Symposium on

Conference_Location

Funchal

Type

conf

DOI

10.1109/ISCC.2014.6912458

Filename

6912458