Content Based Video Retrieval

Abstract: For fast learning and because of availability of fast techniques e-learning and e lecturing concept has more popular. On Internet lots of lectures are available for single concept there an efficient way to perform video retrieval is needed. Video retrieval is a young field which has its genealogy rooted in Artificial Intelligence, Digital Signal Processing, etc, this paper presents a text based video retrieval and video search system. Very first apply automatic video segmentation and key frame detection for video content navigation. Extract textual metadata by applying video OCR technique on key frames and ASR on lecture audio track. This summery will then be used for indexing of video. As there is strong needs for segmenting lecture video into topics for browsing.

Keywords: Content based video retrieval, Optical character recognition, Automatic speech recognition, Segmentation.