Text Based Video Indexing for a Way-Back Machine for Television News
|Title||Text Based Video Indexing for a Way-Back Machine for Television News|
|Publication Type||Conference Proceedings|
|Year of Conference||2013|
|Authors||Goutham PRavi., Sandesh JBananki., Srinivasa G.|
|Conference Name||International Conference on Pattern Recognition Applications and Techniques|
|Keywords||Department of Computer Science and Engineering|
This paper presents a new approach to the problem of retrieving related news videos on the internet for viewing after they have been uploaded. The caption text that appears in most broadcast news videos is one of the major facilitators of content-based retrieval. The caption serves as annotation to the news story. We present a method to extract this text automatically from video clippings of news channels. By grouping related annotations and associating them to each news story, we retrieve news stories efficiently. Thus, the proposed method automates the process of indexing news stories within a video. Thereafter, contextually related video clippings can be stitched to create a timeline to chronologically summarize the news story over a period of time.