CAPIRE: a context-aware points of interest REcognition system using a CBIR approach

  • Authors:
  • Irene Minetti;Silvana Dellepiane;Massimo Valla

  • Affiliations:
  • Department of Biophysical and Electronic Engineering, University of Genoa, Genoa, Italy;Department of Biophysical and Electronic Engineering, University of Genoa, Genoa, Italy;Telecom Italia LAB, Torino, Italy

  • Venue:
  • FMN'10 Proceedings of the Third international conference on Future Multimedia Networking
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes CAPIRE, a service for Points Of Interest (POI) recognition from mobile user-generated photos to provide relevant touristic information. The goal is achieved through the combination of positioning information with image processing and Content-Based Image Retrieval (CBIR) techniques. The system shows to be flexible and fast in learning new classes and robust when more instances of known classes are added to the reference database. No a-priori information or model is considered, but only user-generated photos, which are incrementally added to the knowledge base.