Integration of image, audio, and language understanding for video characterization and variable rate skimming

Digital video is rapidly becoming important for education, entertainment, and a host of multimedia applications. With the size of the video collections growing to thousands of hours, technology is needed to effectively browse segments in a short time without losing the content of the video. We propose a method to extract the significant audio and video information and create a skim video which represents a very short synopsis of the original. This requires integrating language and image understanding techniques for extraction of significant information, such as specific objects, audio keywords and relevant video structure. The resulting skim video is much shorter, and yet retains the essential content of the original segment.;I have developed a system for browsing and previewing digital video, that automatically creates a smaller "skim" video, which is a compact, content rich version of the original. Through the integration of characterization techniques in image and language understanding, subsections of a video are extracted and assembled to create the skim. We have conducted user-studies to test the content summarization and effectiveness of the skim as a browsing tool in a video library.;With increased computing power and electronic storage capacity, the potential for large digital video libraries is growing rapidly. These libraries, such as the Informedia...