Symeon Papadopoulos
Research Engineer, Image clustering, Application development
papadop [AT] iti [DOT] gr
Christos Zigkolis
Research Assistant, Image clustering
chzigkol [AT] iti [DOT] gr
Stefanos Kapiris
MSc Student, Web development
skapiris [AT] csd [DOT] auth [DOT] gr
Yiannis Kompatsiaris
Senior Researcher, Image Analysis, Social Media
ikom [AT] iti [DOT] gr
Athena Vakali
Associate Professor, Web data clustering
avakali [AT] csd [DOT] auth [DOT] gr
Technology
The main goal of ClustTour is to enable tourists to explore new places by use of an interactive map and photo clusters centered on Points of Interest. The application relieves tourists from the information overload commonly present in tourist information sites and tries to answer a simple question: "What are the interesting things to see and do in the place I would like to visit?"
The application is extensively based on the automatic collection, analysis and organization of photographic content from social media sources (currently flickr) and association of the content with structured information (Wikipedia, OpenStreetMap). The massive amounts of photographic content are clustered based on the visual and tag similarities between photos. Eventually, photo clusters are selected that correspond to Points of Interest and they are projected on the map. It is easy then for tourists to explore a place by viewing photos around its Points of Interest.
ClustTour wants to become the first source of information that people will consult when planning their trips and during their trips. It will achieve this thanks to (a) its simple and intuitive interface for exploring a place, (b) interesting content, which will not be limited to the well-known landmarks of a place, but will also contain many niche Points of Interest (thanks to the fact that the content comes from Social Media).
Video
Publications
Symeon Papadopoulos, Christos Zigkolis, Giorgos Tolias, Yannis Kalantidis, Phivos Mylonas, Yiannis Kompatsiaris, Athena Vakali. "Image Clustering through Community Detection on Hybrid Image Similarity Graphs". In ICIP 2010
Symeon Papadopoulos, Yiannis Kompatsiaris, Athena Vakali. "A Graph-based Clustering Scheme for Identifying Related Tags in Folksonomies" In DaWaK 2010
