The detection of scene features in Flickr

  • Authors:

  • Chunjie Zhou;Pengfei Dai;Jianxun Liu

  • Affiliations:

  • School of Information, Renmin University of China, Beijing, China;School of Software, Beijing University of Posts and Telecommunications, Beijing, China;Hunan Knowledge Grid Lab, Hunan University of Science and Technology, China

  • Venue:
  • ICWL'10 Proceedings of the 2010 international conference on New horizons in web-based learning
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Detecting events from web resources has attracted increasing research interests in recent years. Flickr is one of Web resources, which is used to share photos. Complex event detection on Flickr includes the detection of tourist features, user's interest, and so on. With the increasing user requirements of efficient and personalized services, the detection of scene features in Flickr is urgently needed. In this paper we propose a novel method to detect tourist features of every scene, and its difference in different seasons as a probabilistic combination of tags. The use of topic models enables the automatic detection of such patterns, which can translate unstructured tag information into structured event form. The experimental evaluation using real datasets in Flickr show the feasibility and efficiency of the proposed method.