2010 | OriginalPaper | Chapter
Annotating Flickr Photos by Manifold-Ranking Based Tag Ranking and Tag Expanding
Authors : Zheng Liu, Hua Yan
Published in: Information Computing and Applications
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
This paper presents a novel automatic Flickr photos annotation method by ranking user-supplied tags and expanding the top ranked user-supplied tags. Firstly, user-supplied tags are filtered to obtain initial tags by noisy tags pruning. Secondly, the initial tags are ranked using manifold-ranking algorithm. In manifold-ranking process, the photo to be annotated is divided into several regions, and then these regions are acted as queries to launch the manifold-ranking algorithm which ranks the initial tags according to their relevance to the queries. Next, using Flickr API, top ranked initial annotations are expanded by a weighted voting policy. Finally, we combine top ranked initial tags with expanding tags to construct final annotations. Experiments conducted on Flickr photos demonstrate the effectiveness of the proposed approach.