Automatic Video Segmentation and Story-Based Authoring in E-Learning

1 corresponding author: Chun-Hong Huang Abstract— In e-learning, video is widely used in many course domains. We address the problem of the lecture recording and the organization of visual information through user’s interaction at different steps. Our work focuses on the following three important areas: (1) synchronization method of the material contents and (2) interactions between material and users (3) construction of adaptive presentation methods in solving different knowledge levels. This paper proposes a story-based editing and browsing system with the automatic video segmentation. We also point out that a video classification technology can be further integrated to enhance the tool by using visual and audio information. In addition to the semantic segmentation, an instructional video can be edited with an instructor’s story. The story-based editing is similar to hypervideo. Hypervideo is used as a hyperlink in a web. An instructor can constructs an instructional material by hypervideo links. For delivering pre-recorded lectures, we start our discussion on a multimedia presentation recording system which we had developed.