PAPER REVIEWED – Content-based Browsing in Large News Video Databases
This paper particularly derives the effectiveness of content-based browsing paradigm in video retrieval and compared it to the traditional approach of content-based querying with relevance feedback. The model, cluster-temporal browsing, is presented to have integrates feature clusters and chronological video structure dynamically in a single browsing view. This paper contributes into the comparison of cluster temporal browser with a common content-based query paradigm and evaluation of the effect of various browser parameters in search performance. The main conclusion of the paper is that the cluster-temporal browsing has successfully overcome the traditional search paradigm in semantic search topics.
The large amount of video data makes it a tedious and hard job to browse and annotate them by just fast forward and rewind. Recent works in video parsing provide a foundation for building interactive and content based video browsing systems 1,2. In the parsing process, video data are segmented into shots, and each shot may be represented visually by one or more key frames extracted from it. These key frames, together with temporal information of the video shots, can be used for video browsing purpose. If these key frames are extracted automatically in a content based manner3, not just simple sub sampling, key frame browsing could obtain an overview of video data without losing any important information. However, without proper organization, key frames can only be viewed in a sequential looking up manner.
How to provide efficient video browsing facilities has been addressed by many researchers. Generally, a knowledge model is needed to guide the organization of video according to semantic or syntactic contents4. Automatic parsing of news video programs based on syntactic model is a good example of such methods5, where domain model is used to locate and...