2012 | OriginalPaper | Chapter
Caption Text and Keyframe Based Video Retrieval System
Authors : Dung Mai, Kiem Hoang
Published in: Computational Collective Intelligence. Technologies and Applications
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
In this paper, we present a framework for video retrieval using caption text and keyframe similarity. To extract caption text, we applied methods detecting and extracting image areas contain caption text and we used Tesseract-OCR engine to convert into plain text, then use Hunspell library for spell words. Next, we used Clucene search engine index and query on this text. We applied shape descriptors APR and ECM to descript keyframes of the video shots and use those descriptors as a feature vector of video shots. From the feature vectors were obtained, we used ANN library to index and search. The system which is built on the web-based application using ASP.NET support keyword-based and keyframe-based query. The results obtained from experiments produced very promising.