NoteVideo

Abstract

Khan Academy's pre-recorded blackboard-style lecture videos attract millions of online users every month. However, current video navigation tools do not adequately support the kinds of goals that students typically have, like quickly finding a particular concept in a blackboard-style lecture video. This paper reports on the development and evaluation of the new NoteVideo and its improved version, NoteVideo+, systems for identifying the conceptual 'objects' of a blackboard-based video - and then creating a summarized image of the video and using it as an in-scene navigation interface that allows users to directly jump to the video frame where that object first appeared instead of navigating it linearly through time. The research consisted of iteratively implementing the system and then having users perform four different navigation tasks using three different interfaces: Scrubbing, Transcript, and NoteVideo. Results of the study show that participants perform significantly better on all four tasks while using the NoteVideo and its improved version - NoteVideo+ - as compared to others.