Automated analysis and indexing of lecture videos

Abstract

Learning from online videos mainly helps the students and every individual understand a specific topic easily because of the realistic picturization. One of resources available to students is automated analysis and indexing of online lecture videos using image processing. Many online educational organizations and universities use video lectures to support teaching and learning. In past decades, video lecture portals have been widely used and are very popular. The text displayed in these video lectures are a valuable source for analyzing and indexing the lecture contents. Considering this scenario, we present an approach for automatic analysis and indexing of lecture videos using OCR (Optical Character Recognition) technology. For this, we segregated the unique key frames from a lecture video to extract the video contents. After the segregation of key frames by applying OCR and ASR (Automatic Speech Recognition) technology we can extract the textual data contents from the video lecture. From the obtained metadata, we segmented the video lecture based on the time-based text occurrence of the topics. The performance and the effectiveness of proposed analysis and indexing is proven by the evaluation.