| Speaker | Dan Goldman |
| Date | December 1, 2006 |
| Time | 3:00PM to 4:00PM |
| Place | GRAIL (CSE 291) |
Graphical and textual annotations overlaid on a video stream can provide valuable contextual information. Static annotations are easily implemented, but present approaches for dynamic annotations that animate along with objects in the video require either extensive calibration and instrumentation or considerable manual labor. We are developing an interactive authoring system for dynamic annotations on many different types of video objects. Objects in the video are automatically tracked and grouped in an offline preprocess that enables later interactive manipulation. Examples of such annotations include word balloons, captions, titles, graffitti, and editorial notes.