INTRODUCTION

Communication is essential for humans to lead daily life as a member of society. However,
developmentally disabled individuals with motor paralysis such as amyotrophic lateral sclerosis
(ALS) Guillain-Barre-Syndrome or brain-stem infection have difficulty conveying their
intentions because the motor neurons influencing voluntary muscles are affected. Various
assistive technologies that support individual communication have been developed for disabled
people based on a brain-computer interface. Some have facilitated communication with others by
supplementing their impaired functions with surviving functions. In cases of terminal ALS
patients, the eye movement muscles are not typically affected. Human eyesight is one of
mankinds most major senses. The eye is different from the other body parts that make up the
humans sensor array (see Shaviv, 2002). A persons eyes convey a great deal of information with
regards to the meaning behind certain facial expressions. Also, the direction in which an
individual is looking shows where his or her attention is focused. By tracking the position of the
irises, useful interfaces can be developed that allow the user to control and manipulate devices in
a more natural manner. Traditional interfaces are typically one-sided, with the bandwidth from
computer to user far greater than that from user to computer. The movement of a users eyes can
provide a convenient, natural, and high-bandwidth source of additional user input, to help redress
this imbalance (Grauman et. al., 2003 and Mageeet. al., 2004). Some people are so severely
paralyzed that they only have the ability to control the muscles in their eyes. For these people,
eye movements or blinks are the only way to communicate. This research aims in developing a
system that can aid the physically challenged by allowing them to interact with a computer
system using only their eyes.
ABSTRACT
This research is aimed at designing and implementing a human computer interface system that
tracks the direction of the human gaze. The motion and direction of the iris is used to drive the
interface by positioning the mouse cursor accordingly. The location of the iris is done in batch
mode. This implies that the frames are stored in a permanent storage device (hard disk, USB
drive etc) and are retrieved one by one. Each of the frames is processed thus finding the location
of the iris and thereby placing the mouse cursor accordingly. Such a system that detects the iris
position from still images provides an alternate input modality to facilitate computer users with
severe disabilities.
Keywords: Iris Detection, Gaze Tracking.
EYE MOVEMENT-BASED INTERACTION IN VIRTUAL REALITY
Eye movement-based interaction is an example of the emerging non-command based interaction
style. In this type of interaction, the computer observes and interprets user actions instead of
waiting for explicit commands. Interactions become more natural and easier to use. One system
that suggests such advantages is a screen-based system developed by Starker and Bolt. It

monitors eye movements of the user, interprets which objects attract the users interest, and
responds with narration about the selected objects. It minimizes the physical effort required to
interact with the system and increases interactivity. High interactivity is even more important in
VR applications where users often deal with more dynamic and complex environments. The
overall approach in designing eye movement-based interaction techniques is, where possible, to
obtain information from a user's natural eye movements while viewing the display, rather than
requiring the user to make specific trained eye movements to actuate the system. This approach
fits particularly well with virtual reality interaction, because the essence of a VR interface is that
it exploits the user's pre-existing abilities and expectations. Navigating through a conventional
computer system requires a set of learned, unnatural commands, such as keywords to be typed in,
or function keys to be pressed. Navigating through a virtual environment exploits the user's
existing navigational commands, such as positioning his or her head and eyes, turning his or
her body, or walking toward something of interest. By exploiting skills that the user already
possesses, Eye movement based interfaces hold out the promise of reducing the cognitive burden
of interacting with a computer by making such interactions more like interacting with the rest of
the world. An approach to eye movement interaction that relies upon natural eye movements as a
source of user input extends this philosophy. Here, too, the goal is to exploit more of the user's
pre-existing abilities to perform interactions with the computer. Another reason that eye tracking
may be a particularly good match for VR is found in Sibert and Jacob's study on direct
manipulation style user interfaces. They found eye movement-based interaction to be faster than
interaction with the mouse, especially in distant regions. The eye movement-based object
selection task was not well modeled by Fitts Law, or, equivalently, that the Fitts' Law model for
the eye would have a very small slope. That is, the time required to move the eye is only slightly
related to the distance to be moved. This suggests eye gaze interaction will be particularly
beneficial when users need to interact with distant objects, and this is often the case in a virtual
environment. Finally, combining the eye tracker hardware with the head mounted display allows
using the more robust head mounted eye tracker without the inconvenience usually associated
with that type of eye tracker. For many applications, the head-mounted camera assembly, while
not heavy, is much more awkward to use than the remote configuration. However, in a virtual
environment display, if the user is already wearing a head-mounted display device, the headmounted eye tracker adds very little extra weight or complexity. The eye tracker camera obtains
its view of the eye through a beam splitter, without obscuring any part of the user's field of view.
In this study, our first goal was to test the hypothesis that eye movement-based interactions
would perform better in virtual environments than other natural interaction types. In order to test
it, we need to compare against a more conventional interaction technique as a yardstick. We
used hand movement for comparison, to resemble pointing or grabbing interaction that would
commonly be found in virtual environments today. In addition, we investigated whether there
would be performance differences between close and distant virtual environments, i.e.,
where objects are respectively within and beyond the reach of the user. Since pointing-based
interaction requires the user to use hand and arm movements, the user has to move forward in
order to reach and select the objects in the distant virtual environment. Eye movement-based

interaction, however, allows the user to interact with objects naturally using only eye movements
in both close and distant virtual environments. Therefore, we expected eye movement-based
interactions to be faster especially in distant virtual environment.
Eye movement as Inputs
Fixation tracing algorithms take three inputs: eye movement data, target areas, and a process
model grammar.
The eye-movement data comprise sampled points of the form <x, y, v>, where x and y indicate
the location of the point and v indicates the velocity at that point (calculated as point-to-point
distances). The target areas include the name and location of possible fixation targets on the
experiment screen; for instance, for the eye-typing task, the targets would include the letters on
the keypad, the word target that shows the word to be typed and the out target where users look
to signal the end of a trial. The process model grammar represents the cognitive steps undertaken
in a task and the eye movements generated in the execution of these steps. The grammar may be
written directly or derived from models implemented in other systems, such as ACT-R or
GOMS. The grammar comprises regular production rules where the left-hand side contains a
non-terminal (placeholder) and the right-hand side contains a sequence of terminals (symbols)
followed optionally by a non-terminal; the non-terminals represent cognitive subgoals and the
terminals represent target fixations. For instance, Table 1 contains a sample process model
grammar for eye typing one of the words RAT, TRAP, or PART. The model first fires the rule for
the subgoal start, generates a fixation on the target area word, and proceeds to one of the word
subgoals rat, trap, or part. For the word subgoals, the model fixates the letters for the word and
moves to the subgoal end. Finally, the model fixates the out target area, thus ending the trial.
The rules in the grammar include probabilities to model the likelihood of each rule firing for a
particular subgoal; for the sample grammar, the rules for the start subgoal are equally likely.
Fixation tracing produces two outputs: a model trace and a model evaluation. The model trace
represents a mapping from eye movement data points to the fixation sequence predicted by the
best corresponding model strategy. The model evaluation represents the probability of the model
trace, which can be used to evaluate the fit of the model to the data. In this paper, we will use
only the model trace and ignore the model evaluation. However, future work on eye-based
interfaces could utilize the model evaluation to determine the likelihood that its interpretation of
observed eye movements is correct.
Eye Tracking Devices
Recently, there has been a trend to use standard camera components such as a webcam or mobile
cam to do some of what an eye tracker can do.
Standard webcam technology is used to determine the more indicative studies of visual
attention and direction of interest of website users. Not as precise as standard eye tracking, but

provides a general and cost-effective solution to non-scientific studies for online websites and
other visuals provided on a computer.
Mobile camera technology is used in some smartphones for eye presence detection. It is
primarily a power-saving feature and, contrary to eye tracking technology with illumination, you
are limited by the ambient light in the environment

CONCLUSION
An eye tracker as an input device is far from perfect, in the sense that a mouse or keyboard is,
and that is caused both by the limitations of current equipment and, more importantly, by the
nature of human eye movements. Accuracy obtainable is more similar to a traditional touch
screen than a mouse, and the range can barely cover a single CRT display. The equipment, while
non-intrusive and non-contacting, is difcult to ignore. Nevertheless, it is perhaps amazing that
eye movement-based interaction can be done at all; when the system is working well, it can give
the powerful impression of responding to its users intentions rather than his or her explicit
inputs. To achieve this, our overall approach in designing interaction techniques is, wherever
possible, to obtain information from a users natural eye movements while viewing the screen
rather than requiring the user to make specic eye movements to actuate the system. For
example, we tried and rejected long gazes because they are not natural eye movements,
preferring to use gazes only as long as natural xations. We also found it important to search forand recognize xations in the raw eye tracker data stream and construct our dialogue around
these higher-level events. We saw that designing interaction techniques around a philosophy that
emphasizes exploiting natural eye movements meshes well with the virtual environment
interaction style, which is also based on natural navigational commands. Eye movement-based
interaction and virtual environment interaction also share the non-command-based property,
which will increasingly characterize advanced user-computer interfaces. And, in both areas,
working from the natural characteristics of the human users to the interface designs leads to
powerful improvements in the naturalness, convenience, and, ultimately, performance obtainable
from advanced interface technologies. Finally, we saw how eye movement-based interaction and
virtual environments are instances of an emerging new style of user-computer interaction. This
style is characterized by a change from explicit to implicit commands as well as a change from
turn-taking, singlestream dialogues to simultaneous, parallel interactions.

REFERENCE
http://pdf.aminer.org/000/090/151/inferring_intent_in_eye_based_interfaces_tracing_eye_move
ments_with.pdf
Human Computer Interaction Third Edition Dix Finally Abowed Beale By ALAN DIX, JANET
FINALY GREGORY D. ABOWD, RUSSELL BEALE.
. R.A. Bolt, Gaze-Orchestrated Dynamic Windows, Computer Graphics 15(3) pp. 109-119
(August 1981).- 37 R.A. Bolt, Eyes at the Interface, Proc. ACM Human Factors in Computer Systems
Conference pp. 360-362 (1982).
H.D. Crane and C.M. Steele, Generation-V Dual-Purkinje-image Eyetracker, Applied Optics
24(4) pp. 527-537 (1985).
B.N. Flagg, Children and Television: Effects of Stimulus Repetition on Eye Activity,
Thesis, Doctor of Education degree, Graduate School of Education, Harvard University
(1977).
J.D. Foley, A. van Dam, S.K. Feiner, and J.F. Hughes, Computer Graphics: Principles
and Practice, Addison-Wesley, Reading, Mass. (1990).
F.A. Glenn and others, Eye-voice-controlled Interface, Proc. 30th Annual Meeting of
the Human Factors Society pp. 322-326, Santa Monica, Calif. (1986).
M. Green and R.J.K. Jacob, Software Architectures and Metaphors for Non-WIMP User
Interfaces, Computer Graphics 25(3) pp. 229-235 (July 1991).
R.N. Haber and M. Hershenson, The Psychology of Visual Perception, Holt, Rinehart and
Winston, New York (1973).
T.E. Hutchinson, K.P. White, W.N. Martin, K.C. Reichert, and L.A. Frey, HumanComputer
Interaction Using Eye-Gaze Input, IEEE Transactions on Systems, Man, and Cybernetics 19(6)
pp. 1527-1534 (1989).
R.J.K. Jacob, A Specication Language for Direct Manipulation User Interfaces, ACM
Transactions on Graphics 5(4) pp. 283-317 (1986).
http://www.cs.tufts.edu/jacob/papers/tog.txt [ASCII];