Integration of multimodal multimedia database system architecture with query level fusion

Thesis Type: Postgraduate

Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Faculty of Engineering, Department of Computer Engineering, Turkey

Approval Date: 2013


Consultant: ADNAN YAZICI


Multimedia data particularly digital videos that contain various modalities (visual, audio, and text) are complex and time consuming to deal with. Therefore, managing large volume of multimedia data reveals the necessity for efficient methods of modeling, processing, storing and retrieving these data. In this study, we investigate some of the requirements to efficiently deal with multimedia data, especially video data. To satisfy such requirements we aim to integrate specific multimedia database architecture which consists of semantic content extractor, storage, index, query and coordinator modules. In addition, to simplify complicated and time consuming operations on video files some client-server based applications with appropriate graphical user interfaces are implemented to carry out these operations in multi-threaded mode. The proposed architecture also supports different query types including the combination of content as well as concept-based queries which provides users with the ability to perform multimodal query. Furthermore, we introduce a fusion approach at the query level to improve query retrieval performance of the multimedia database system.