Archive notice: This article was originally published on July 31, 2006. Links and embedded videos are preserved as part of the historical record.
The four authors Dan B. Goldman, Brian Curless, Steven M. Seitz and David Salesin showed in their Siggraph paper “Schematic Storyboarding for Video Visualization and Editing” a way to generate storyboards from video material that follow the visual language of painted storyboards and are more informative than stills.
In other words, the method shown in this paper generates a storyboard from short video sequences that is supplemented with directional cues. Compared with individual images, movement relationships become clear here and periods of time can be grasped intuitively. This makes the technology suitable for simplifying video editing, for analyzing surveillance data or classic films, or even for generating comics from film sources.
The technology automatically detects camera pans and zooms, combines the associated individual images into overview panoramas and annotates them using arrows.
For people, the technology analyzes changes in position and posture, isolates key images and then inserts them into the camera panoramas. Graphic symbols then establish the connection.
Illustrative examples and details can be found on the authors’ website.
Further links:
- Schematic Storyboarding
