Detail-on-Demand Hypervideo John Doherty, Andreas Girgensohn, Jonathan Helfman, Frank Shipman, Lynn Wilcox FX Palo Alto Laboratory 3400 Hillview Avenue Bldg. 4 Palo Alto, CA 94304 1-650-813-7574 wilcox@fxpal.com ABSTRACT We demonstrate the use of detail-on-demand hypervideo in interactive training and video summarization. Detail-on-demand video allows viewers to watch short video segments and to follow hyperlinks to see additional detail. The player for detail-on- demand video displays keyframes indicating what links are available at each point in the video. The Hyper-Hitchcock authoring tool helps users create hypervideo by automatically dividing video into clips that can be combined in a direct manipulation interface. Clips can be grouped into composites and hyperlinks can be placed between clips and composites. A summarization algorithm creates multi-level hypervideo summaries from linear video by automatically selecting clips and placing links between them. Categories and Subject Descriptors H.5.1 [Information Interfaces and Presentation]: Multimedia Information Systems – video. H.5.4 [Information Interfaces and Presentation]: Hypertext/Hypermedia – navigation, user issues. General Terms Algorithms, Design, Experimentation, Human Factors. Keywords Hypervideo, video summarization, link generation, video editing. 1. INTRODUCTION Detail-on-demand [1] video is a type of hypervideo where a single button press reveals additional material related to the currently playing video. This is similar to certain DVDs that provide links to other video. When a link is active, an icon appears on top of the playing video. The user can push a button to jump to the alternate video. For example, in The Matrix and other movies, links take the viewer to video segments explaining how the scene was filmed. Afterwards, the original video continues from where the viewer left off. Detail-on-demand video is well suited as training video because it allows users to easily navigate to the parts of interest. The top level of video provides an outline of the material, with links to additional material on the current topic. The user can remain at the top level, or follow a link for more detail. Unlike hierarchical keyframe navigation [2] that allows the user to control playback from a separate window, detail-on-demand video allows the user to navigate between levels of detail while the video is playing. Detail-on-demand video is also suitable for video summarization. In a typical video summarization, the user must decide a level of detail a priori [3]. With detail-on-demand, the user can watch a multi-level video summary that includes summaries of different levels of detail. The user can dynamically control the level of detail by choosing or not choosing links. Figure 1. Detail-on-demand hypervideo player. 2. HYPERVIDEO VIEWING We have developed a player for viewing detail-on-demand video that combines the characteristics of browsing the Web and changing channels on TV. As the viewer watches a video, the player indicates that a link is available from the currently playing clip by showing a keyframe and label for the link in the lower right hand corner of the application, as seen in Figure 1. The timeline of the player also indicates where links are available. It shows the labels for all available links to give the viewer a quick overview of the possible link destinations. The user follows a link by clicking on the keyframe. When a link is followed, the keyframe animates into the video and the video clip corresponding to the link begins to play. At the same time, another keyframe and label appear in the lower left hand corner of Copyright is held by the author/owner(s). MM’03, November 2-8, 2003, Berkeley, California, USA. ACM 1-58113-722-2/03/0011.