1. Introduction 1.1 CVN (Columbia Video Network) CVN tapes video and sound for a number of courses taught at Columbia University. The recorded information is then sent in the form of VHS videotapes to students participating in distance learning, and it is also made available as streaming media for students taking the course on-campus. In the process of recording video, a separate program processes the video data and selects key frames that best summarize the video information for a certain period of time. For a 75 minute class, we can expect around 210 such key frames on average. These key frames prove to be valuable in reviewing a given class by providing a means to search and look at pertinent data from the class. So far, CVN makes this information available through a web interface. The top left portion of the interface allows the user to view the streaming video and audio data. The upper right portion displays the key frame closest to the position in the streaming video. The lower portion of the interface, the one CVNView attempts to improve and build upon, displays the collection of key frames, ordered by their appearance over time.

Figure 1: CVN Interface to Class Videos and Key Frames

1.2 Analysis of the Problem The usability of the static collection of slides is questionable. The interface merely reflects a massive number of slides that can be viewed one-at-a-time. If the intent of a user is to search for some particular section of a video, the user can either watch the entire video, or click through all slides. Assuming further that any given University course is taught twice a week for 13 weeks, the user may have to end up browsing through 26 * 75 minutes of video (1950 minutes = 32.5 hours) or 26 * 210 slides (5460 slides). Moreover, no attempt has yet been made to relate the slides in any way other than ordering them by time. We can expect from most well-organized classes to discuss several topics, possibly in some meaningful order. Each topic is then broken into several sections. As each such section tends to be discussed in hopefully more than 18 or so seconds (the average time between key frames), it may be more useful to compress all key frames for a given section into one summarizing slide. This method will shrink the number of non-repetitive slides per class to perhaps one tenth of the original number. Inarguably it is much more useful to browse 21 pertinent slides, each of which expands into a section or topic, as opposed to 210 repetitive ones. While CVNView was not developed to compress 5460 slides into one usable computer screen, it does provide some insight into the matter of visualizing a large amount of slides and it shows possibilities of relating slides by categorizing them not only by time, but also by section, topic, and location in the classroom.