Determining Organization of Educational Lecture Video Based on Presentation’s Contents

Videos have become a very important part of higher education institutes. It is integrated as part of any traditional courses. Even though video lectures are very interesting teaching resource for the students, but they lack in accessing the required contents. We cannot access to the specified section of video unless the location is known. The main aim of this chapter is to automatically partition an educational lecture video into various topical segments which is then presented to the students in a customized manner for easy navigation. This approach identifies the topics being covered by the instructor, based on text similarities across the various frames in the lecture video. Fundamental problem in any lecture video is to give semantic query and effectively retrieve relevant contents form long video. Effective search capability for the students can be provided if proper browsing facility is available. Our approach concentrates on the slides instead of a presenter from the projected electronic slides. The method aims in using the video streams of the lecture delivered captured from camera and get the slides converted into images after applying slide segmentation. These independent images have heading and subheadings which will be used for organizing the lecture video. The organizing and segmenting educational lecture videos of the technical contents according to slide changes not only structure the content of video according to topics, but also facilitate the synchronization of video, audio and electronic slides for effective indexing, browsing and retrieval at later stage. The purpose of this work is to make learner easy access to the video contents. Getting to a specific page or paragraph in the text books is easy because of the content page in the beginning of the book and index page at the end of the book. In the similar context; lecture video need a system to extract a text from the presentation slides. The main objective of this study is to develop effective and efficient automated approaches to segment lecture videos into topics and subtopics. This video segmentation and browsing structures will support knowledge searching and e-Learning.