Efficient Multimedia Information Retrieval with Query Level Fusion


Sattari S., YAZICI A.

11th International Conference on Flexible Query Answering Systems (FQAS), Cracow, Poland, 26 - 28 October 2015, vol.400, pp.367-379 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 400
  • Doi Number: 10.1007/978-3-319-26154-6_28
  • City: Cracow
  • Country: Poland
  • Page Numbers: pp.367-379

Abstract

Multimedia data particularly digital videos that contain various modalities (visual, audio, and text) are complex and time consuming to deal with. Therefore, managing a large volume of multimedia data reveals the necessity for efficient methods for modeling, processing, storing and retrieving such data. In this study, we investigate how to efficiently manage multimedia data, especially video data. In addition, we discuss various flexible query types including the combination of content as well as concept-based queries that provide users with the ability to perform multimodal query. Furthermore, we introduce a fusion-based approach at the query level to improve query retrieval performance of the multimedia database. Our experimental tests show a significant improvement in the query retrieval performance.