Designing a View Management System for Augmented Reality

wafflejourneyΤεχνίτη Νοημοσύνη και Ρομποτική

14 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

80 εμφανίσεις

P
ROJECT
P
ROPOSAL

Sean Tai

Mentor: Dennis Brown

Firm: Naval Research Lab in Washington, D.C.


I.
Title of the Project
.

Designing a View Management System for Augmented Reality


II.
Problem Statement
. The purpose of this project is to design a theoretical

computer
-
based system for
gathering information
related to

a soldier wearing a mobile Augmented Reality (AR) system, filtering the
information based on relevancy, and then outputting the information in a useful format
suitable
for graphical
rendering. Al
l of these steps are part of View management, which is gathering, managing, and outputting
information in AR.


III.
Purpose
. This project will be used in the code libraries at the Naval Research Lab (NRL) in Washington,
D.C. Scientists developing AR ther
e will be able to use this theoretical/high
-
level programming in their future
work.

Ideally, this designed system will be both feasible and implementable.
The project deals with new and
upcoming technology in the computer science world, and will be helpf
ul to those working in fields such as AR,
Virtual Reality, and Mixed Reality.


IV.
Scope of Study
.
As this project will be almost completely theoretical with possibly some implementation
and coding, it will depend largely on ideas from published papers o
n View Management

in AR as well as from
the AR scientists at NRL.

Much

of the content will be ideas that
branch

off of others’ thinking. If time and
resources permit, a demo of the Management system could
demonstrate

how sample data is handled in an
appl
ication setting.


V.
Background and review of current literature/research in this area
. Kaiser et al.
’s group
(2003)
studied
how to determine what a user is looking and interacting with at a time, which could analyze what is most
important to a person at
any given time.
The researchers

came up with ideas such as “Regions of Interest”
with tracking eyesight and gestures and an object ranking system (e.g., visibility of an object) to see what was
most focused on by the
user
. Romero and Correria (2003) used

a “context mode” system for representing what
a person was doing at a point in time, and incorporated user preferences and past experiences stored in a
database to filter real
-
time information.

Rhodes (2003)
developed a
rule base which
indirectly
incorpo
rated a
time element to determine relevancy

of objects
.

At NRL (Julier, Brown, and Baillot, 2001), a two step system
was used. In the first, objects
deemed

close enough in proximity to the user were chosen. Then, task
-
dependency filtered what was to be
shown.

I
n another instance (Höllerer et al., 2001),
objective and subjective
properties were used to classify objects

for information filtering.


VI.
Procedure and Methodology
. Since a majority of the project is theory
-
related,
I will be mostly working
i
ndependently with ideas that I discover and explore. I will be using ideas that I learned from a design class
that I took at George Mason University
with Dr. Richard Evans
last semester

to design the system with an
emphasis on considering alternatives
, an
d I will also be discussing ideas with
some colleagues, such as fellow
interns. In the three weeks or so, I hope to finish developing a basic design of an information filtering system,
and sometime after that I plan to finish designing more detailed eleme
nts of the system, such as
pseudo
-
code
implementations
.

If time permits, I may try creating an application

in C++

to test out my ideas
, possibly using
the Epsilon 3
-
D engine code base that is being written right now
.

It may start with some data and then
demonstrate information filtering techniques
.


VII. Expected Results and Value to Others
.
I expect the final product, a theoretical design of a View
Management system which may or may not include an application,
to be a comprehensive idea that researchers

at NRL could take into account and even turn into a full
-
working application.

This should be done by the time
the SEAP internship ends on August 18.
The project could be presented
with illustrations of key ideas,
such as

the information filtering system
.

I will probably use a
“decision time

line” format from
said
taken
design
course to plan and chart my progress.

References


Hollerer, T., Feiner, S., Hallaway, D., Bell, B., Lanzagorta, M., Brown, D., et al. (2001, October). User
interface management te
chniques for collaborative mobile augmented reality.
Computers and
Graphics, 25
(5), 799
-
810.

Julier, S., Brown, D., & Baillot, Y. (2001, April). The need for AI: intuitive user interfaces for mobile
augmented reality systems.
Unknown
.

Kaiser, E., Olwal, A.
, McGee, D., Benko, H., Corradini, A., Li, X., et al. (2003). Mutual disambiguation of 3D
multimodal interaction in augmented and virtual reality.
Proceedings of the 5th International
Conference on Multimodal Interfaces
, 12
-
19. Retrieved July 3, 2006, from

http://portal
.acm.org/
citation.cfm?id=958438&coll=GUIDE&dl=ACM&CFID=11111111&CFTOKEN=2222222&ret=1#F
ulltext

Rhodes, B. (2003, August). Using physical context for Just
-
in
-
Time information retrieval.
IEEE Transactions
on Computers
. Retrieved July 3, 2006, from http://ieeexplore.ieee.org/
iel5/
12/
27484/
01223636.pdf?tp=&arnumber=1223636&isnumber=27484

Romero, L., & Correia, N. (2003). HyperReal: A Hypermedia model for mixed reality.
Proceedings of the
Fourteenth ACM Conference on
Hypertext and Hypermedia
, 2
-
9. Retrieved July 3, 2006, from
http://delivery.acm.org/
10.1145/
910000/
900055/
p2
-
romero.pdf?key1=900055&key2=9238251511&coll=GUIDE&dl=portal,ACM&CFID=11111111&C
FTOKEN=2222222