Unsupervised Clustering of Social Events

Unsupervised Clustering of Social Events

Abstract

This paper describes our contribution to the social event detection (SED) task of the MediaEval Benchmark 2013. We present a robust unsupervised approach for the clustering of tagged photos and videos into social events. Results on the SED datasets show that the proposed approach yields an excellent generalization ability and state-of-the-art clustering performance.

Grafik Top
Authors
  • Zeppelzauer, Matthias
  • Zaharieva, Maia
  • Del Fabro, Manfred
Grafik Top
Shortfacts
Category
Paper in Conference Proceedings or in Workshop Proceedings (Paper)
Event Title
MediaEval 2013 Workshop
Divisions
Multimedia Information Systems
Event Location
Barcelona, Spain
Event Type
Workshop
Event Dates
October 18-19, 2013
Publisher
CEUR-WS.org
Date
2013
Official URL
http://ceur-ws.org/Vol-1043/
Export
Grafik Top