Thursday, May 8, 2003 - 3:30, NSH 3001
Negative Pseudo Relevance Feedback for Multimedia Retrieval
Rong Yan

Abstract:
Multimedia retrieval requires a system to find information relevant to a query which may be presented simultaneously in different ways through a text description, audio, still images and/or video sequences. One of the important issues in multimedia retrieval is to select the proper similarity scores during the retrieval process. In this study, we propose a negative pseudo relevance feedback approach that uses the retrieved items that are NOT similar to the relevant items without further inquiring the user feedback. We provide insight into this approach using a statistical model and suggest a score combination scheme via posterior probability estimation. An evaluation on the 2002 TREC Video Track queries is used to demonstrate the effectiveness of this approach.

Related Readings:
Boosting Image Retrieval
K. Tieu, P. Voila
In Proceedings of IEEE Computer Vision and Pattern Recognition 2000

On the Choice of Similarity Measures for Image Retrieval by Example
J. P. Tarel, S. Boughorbel
In Proceedings of ACM Multimedia 2002

A Survey on the Use of Pattern Recognition Methods for Abstraction, Indexing, and Retrieval of Images and Video
S. Antania, R. Kasturi, R. Jain
In Pattern Recognition, vol. 4, 2002