thesis_document.pdf

Projector-Based Location Discovery and Tracking

Johnny Chung Lee

May 2008

CMU-HCII-08-102

Human Computer Interaction Institute

School of Computer Science

Carnegie Mellon University

Pittsburgh, PA 15213

Thesis Committee:

Scott E. Hudson (Chair), Carnegie Mellon University

Jodi Forlizzi, Carnegie Mellon University

Luis Von Ahn, Carnegie Mellon University

Paul Dietz, Mitsubishi Electric Research Labs

Submitted in partial fulfillment of the requirements for

the degree of Doctor of Philosophy

This research was supported in part by the National Science Foundation under grants

CCR-03244770, IIS-0121560, IIS-0205219, IIS-0325351, IIS-0329090, IIS-9800597, IIS-

9980013, Mitsubishi Electric Research Labs, Intel Research Council and General Motors.

Keywords: Image projection, projector calibration, augmented reality, location

tracking, motion capture, high-speed projection, infrared projection, multi-projector

applications, projector alignment, incremental tracking.

Abstract

Today, the primary use of projection technology is for creating large flat displays

that provide a shared viewing experience for presentations or entertainment

applications. While research projects have explored the powerful ability for

projected light to create illusions that can reshape our perception and our

interaction with surfaces in the environment, very few of these systems have had

success in terms of commercial and consumer adoption. Part of this limited

adoption can be attributed to the lack of practicality in the cost-of-operation due

to the complexity of installation and reliability of execution. Often these systems

require expert knowledge to perform system setup and calibration between the

projected image and the physical surfaces to make these illusions effective. In this

thesis, I present a technique for inherently adding object location discovery and

tracking capabilities to commercial projectors. This is accomplished by

introducing light sensors into the projection area and then spatially encoding the

image area using a series of structured light patterns. This delivers a unique

pattern of light to every pixel in the projector’s screen space directly encoding the

location data using the projector itself.

By unifying the image projection and location tracking technologies, many of the

difficult calibration and alignment issues related to interactive projection and

projected spatial augmented reality applications can be eliminated simplifying

their implementation and execution. Furthermore, by creating a hybrid visible

light and infrared light projector, a single calibration-free device can perform

invisible location tracking of input devices while simultaneously presenting

visible application content. I present a detailed description of the projector-based

location discovery and tracking technique, a description of three prototype

implementations, and a demonstration of the effectiveness of this simplification

by re-implementing, and in some cases improving upon, several location-sensitive

projector applications that have been previously executed using external

calibration and tracking technologies.

iii

Acknowledgements

The primary influences in this work most certainly are my thesis advisor, Scott

Hudson, and research mentor, Paul Dietz.

Scott provided the research guidance and outlook that has taken me from a

naïve new graduate student to the researcher I have become today. When I began

the doctoral program at Carnegie Mellon University, I had little knowledge and

appreciation of what research should be and a lack of confidence that I could

make a meaningful contribution to the field. Scott has been integral in helping me

gain perspective, to select problems of substance, and taught me the approach to

develop solutions that are practical. These two skills have become essential in

defining my path of work as a researcher.

Paul provided me with an exposure to a sense of technical creativity and

personal enjoyment in invention that I have not found in any other individual. As

mentor, he has provided me with a perspective on life and the impact that work,

research, and creativity can have on the world outside the typical academic

boundaries. For this, I will be eternally grateful

In addition to these two individuals, I have had the pleasure to work with

and be in the company of brilliant individuals that have inspired me on both

professional and personal levels. This includes my co-authors Ramesh Raskar,

Desney Tan, Dan Maynes-Aminzade, Jay Summit, Chris Kyriakakis, and Darren

Leigh. My classmates, colleagues, and friends James Fogarty, Daniel Avrahami,

Darren Gergle, Luis Von Ahn, and Laura Dabbish have been an incomparable set

of individuals to have begun my research career with. I felt sad to see them

graduate and move on, but I now look forward to joining them “on the other side”.

My brother, David Lee, was the one who first got me interested in Human-

Computer Interaction when he was taking classes at University of Virginia with

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: