American Sign Language is a language different from English, with its own unique syntax and semantics. It is at least as different from English as any other spoken foreign language. An automatic English to ASL translator would give the Deaf greater access to the English-speaking world.
The current emphasis of the project is in developing the tools to generate ASL signs as animation. This poses additional challenges to those of conventional animation, where a design specifies a set of motions that are carried out along a single timeline. Here the goal is to develop transitions between sequences of signs that will be acceptable to the Deaf. In particular, we are looking at ways to avoid the return to a neutral position between signs and nonmanual signals. For more information, visit DePaul's ASL Web Site
Mary Jo Davidson, Jorge Toro, Karen Alkoby, Jerry Schnepp and Brent Shiver completed their dissertations in this area of inquiry.
A critical skill for success in a computer graphics career is a well-developed visual sense or an "eye". No matter how many technical skills are listed on a resume, a computer graphics firm will want to see a portfolio to judge the visual ability of the applicant. Unfortunately many computer scientists are uncomfortable with the idea of straying from the math, physics and data structures of computer graphics in order to teach a skill involving observing and describing an image.
This problem lead to the development of a pedagogical technique called visual analysis, which is an easy-to-use method to help students analyze an image for rendering algorithms. In addition to learning characterize a rendering algorithm in terms of its time and space requirements, students learn to characterize the visual behavior of an algorithms. At present this effort has resulted in several papers, supporting software and a book
A visual demonstration of structure or behavior can be a powerful tool for study in that it can provide a picture to bolster textual explanations. Some challenges involve the display of information drawn from noisy data and the development of imagery to aid the comprehension of algorithms.
Quantization can be used as a method of image compression. Color quantization is the process of representing a full-color image with selected subset of representative colors. An image containing fewer colors takes up less space and requires less time to transmit.
Interestingly, no one undertook studies with humans to determine the perceived image quality of the resulting quantized images until this effort. Dan Kusswurm extended the method from still images to video sequences and completed his PhD at DePaul in 1998.
In a rapidly changing field, the introduction of new tools and techniques must emphasize the enduring fundamentals of the discipline. Seminal Graphics is a juried collection of papers that shaped the field of computer graphics and is a good collection for a student embarking on a career in computer graphics.
Other perennial interests include curricula for computer graphics as well as resources and techniques for teaching various topics in the discipline.
The wonderful thing about graphics is that all that study of art, physics, mathematics, computability and the psychology of human perception can be used to create pictures!
This page contains images that have appeared on various book covers, CD jackets, exhibits and other public walls.
Ready for you
Erik the Red, Green and Blue