Seminar report on

EYE GAZE
An inner vision of emerging technology

By
Yashaswi P.N Aditya Kalluraya III-year EC Yashaswi99@gmail.com adityakalluraya@gmail.com

Department of Electronics and Communication

Vivekananda College of Engg and Technology Puttur 574203

Abstract Eye-gaze is an input mode which has the potential of an efficient computer interface. Eye movement has been the focus of research in this area. Non-intrusive eye gaze tracking that allows slight head movement is addressed in this paper. A small 2D mark is employed as a reference to compensate for this movement. The iris center has been chosen for purposes of measuring eye movement. The gaze point is estimated after acquiring the eye movement data. Eye gaze interaction can provide a convenient and natural addition to user-computer dialogues. Eye gaze interaction technique is faster than selection with a mouse. It is particularly beneficial for the larger screen workspaces and virtual environments of the future, and it will become increasingly practical as eye tracker technology matures.

I.

INTRODUCTION

Human-computer interaction has become an increasingly important part of our daily lives. The movement of users eyes can provide a convenient, natural and high-bandwidth source of input. By tracking the direction of gaze of the user, the bandwidth of communication from the user to the computer can be increased by using the information about what the user is looking at, and even designing objects specially intended for the user to look at. A variety of eye-gaze (eye-movement) tracking techniques exist. A short list includes lectra-Oculography Limbus Pupil and Eyelid Tracking Contact Lens Method Cornea1 and Pupil Reflection Relationship Purkinje Image Tracking Artificial Neural Networks and Head Movement Measurement

Figure 1. head mounted system

II.

TRACKING OF EYE MOVEMENT

The location of face and eye should be known for tracking eye movements. We assume this location information has already been obtained through extant techniques. Exact eye movements can be measured by special techniques. This investigation concentrates on tracking eye movement itself. The primary goal of this paper is to detect the exact eye position. Two algorithms have been proposed for iris center detection: the Longest Line Scanning and Occluded Circular Edge Matching algorithms. The emphasis is on eye movement in this paper, not on face and eye location. Rough eye position is not sufficient for tracking eye gaze accurately. Measuring the direction of visual attention of the eyes requires more precise data from eye image. A distinctive feature of the eye image should be measured in any arrangement. The pupil of people having dark or darkbrown eyes can hardly be differentiated from the iris in the captured images. If the image is captured from close range, then it can be used to detect the pupil even under ordinary lighting conditions. It was decided to track the iris for this reason. Due to the fact that the sclera is light and the iris is dark, this boundary can easily be optically detected and tracked. It can be quite appropriate for people with darker iris color. However, there are some issues, which have to be emphasized. They arise, due to the following reasons: Coverage of the top and bottom of the limbus by the eyelids. Excessive coverage of the eyes by eyelids.

Computer vision is intrinsically non-intrusive, and does not require any overly expensive equipment. Non obtrusive sensing technology - such as video cameras and microphones has received special attention in this regard. This paper draws on computer vision and image processing techniques for measuring eye-gaze. Systems based on different technologies have been implemented, in many cases with some devices mounted on the person head. Also, non-intrusive systems have been developed and are being used in many applications.

The techniques proposed in this section effectively deal with the first A.

front of the screen. This observation justifies a circular approximation to the ellipse. Experimental results justify this simplification. III. HARDWARE SYSTEM

Longest Line Scanning (LLS)

The principle of the IRIS technique is based upon the reflection of IR light by the area on both sides of the edge between the white sclera and the darker iris, which is called the limbus. Two IR light-emitting diodes and IR lightsensitive phototransistors, positioned, respectively, above and below each eye, are used to collect the difference of IR light reflected by the eye. This difference is preprocessed and converted into an analog voltage, which is proportional to the angular deviation of the eye. In all our experiments, the horizontal eye movements (i.e., the x-coordinates) are recorded from the right eye, and the vertical eye movements (i.e., the y-coordinates) are recorded from the left eye. A PC-based system is employed to carry out the acquisition and calibration of the eye movements. It is also employed to graphically display the image and stimulation dots. The PC is placed at 60 cm from the subject under investigation, and the subject sits with his/her head stabilized by an optometric structure and chin rest. The IRIS output signals from each eye are sampled at a frequency of 500 Hz, with the help of a 12-b digital-interface board. After the initial adjustments of the eye tracker are carried out, the subject is asked to look at a series of dots displayed at different locations on the PC screen. On Fig. 2, we can see an example of distortion that affects the measured gaze position that is obtained for each reference dot presented to the subject. In order to overlap the distorted grid, which is obtained from the measured gaze positions, with the grid of reference, it is necessary to transform the grid of gaze position into the same size and projection as the grid of reference. We call this transformation process spatial mapping, and only after this transformation will it be possible to get the accurate gaze position on the screen.

Human eyes have three degrees of freedom of rotation in 3D space. Actually, the eye image is a projection of the real eye. The iris is nearly a circular plate attached to the approximately spherical eyeball. The following well-known property is useful in this regard. The center of an ellipse lies on the center of the longest horizontal line inside the boundary of the ellipse The LLS algorithm is an application of this property. It can be applied to the problem of detection of the iris center. Searching and decision after edge detection enhances computational efficiency. Except when preprocessing fails, it computes the center of the iris quite accurately. But it is sensitive to distribution of edge pixels.

Figure 2. principles of LLS

B.

Occluded Circular Edge Matching(OCEM)

Although the LLS method detects the center of the iris, it is not sufficient for measuring eye-gaze precisely. The following Problems are noted on a closer look at LLS technique intra-iris noise rough iris edge occlusion of longest line by eyelids

The only clues to find the center of the iris are left and right edge pixels of the iris boundary, the so called limbus. In order to estimate the original position and shape of the iris boundary, the circular edge matching (CEM) method can be adapted. As mentioned earlier, the iris is naturally occluded by eyelids to some extent, depending upon the individual or the status of the subject. CEM should be adaptively modified. Only the visible portions of the edge without occluded portions need to be processed in the matching step. The angle of rotation of the eyeball and the eccentricity of the ellipse are not large, when the subject sits and operates the computer in

. Figure 3. Subject positioned for eye-movement recording.

applied to the eye in order to avoid the risk of hurting it, following the ophthalmologic recommendations and working way below the dangerous levels. Each illuminator consists of 20 high intensity infrared LED with aperture angle of 6 and located at an average of 75 cm from the eyes. That enables enough face illumination and the generation of the glints. This illumination is synchronized with the electronic shutter of the NFV camera to generate 500 microseconds long flashes every 20 milliseconds.

Figure 4. Example of distorted-gaze position,where the solid line represents the reference grid,and the dotted line represents the measured gaze position.

IV.

GENERAL SYSTEM OVERVIEW

The system analyses an image of the eye and determines the gaze direction by computing the vector defined by the pupil centre and a set of glints generated in the eye by an infrared illuminator. To optimize the resolution, the vector has to be as large as possible; therefore a very narrow field of view (NFV) is needed in the eye camera. In this situation, to centre the eye in the image, the camera has to be moved to follow the eye and compensate for the head movements. When, due to a sudden head movement, the eye gets out of the NFV camera image, the eye is quickly located using another camera, this one with a wide field of view (WFV). The change in the curvature between iris and sclera can provoke distortions of the glint complicate the determination of its co-ordinates and its geometrical characteristics. The glint can also disappear for strong torsions of the eye. The use of two illuminators to guarantee at least one valid glint is not enough, therefore a configuration with four illuminators has been chosen for this system. In case of having more than one valid glint, they have to be identified. The specific shape of the illuminators allows the optimization of the identification algorithm. With this approach, the centre and radius of the iris has to be determined to reject the glints found outside the iris area. Another problem is the influence of external illumination variations in the quality of the image. To minimize it, the shutter of the NFV camera is set to 1/2000s and a powerful infrared illumination is used. Special care has been taken in limiting the amount of illumination power

Figure 5. System overview

The NFV camera has a 1/2 infrared sensitive CCD with minimum illumination of 0.1 lux. A focal length of 150 mm is obtained with a 75-mm lens and a 2 extender. The F-stop has to be set over 5.6 to allow for enough depth of field. The WFV camera has a 1/3 infrared sensitive CCD with a minimum illumination of 0.5 lux. The system is able to process in real time 25 frames per second for images of 768576 pixels. It permits gaze tracking for vision angles over 45.

V.

APPLICATIONS
Figure 6. HUD from BMW 5 series.

1) EYE GAZE TRACKING IN WEB,IMAGE AND VIDEO DOCUMENTS: Few works have been dedicated to extracting user interest on the basis of gaze patterns. In other words, few works in the literature give a sense to the human-multimedia dialogue. The state of art lacks too on semantics associated to eye gaze/fixation communication between the human and multimedia documents. A wide variety of interdisciplinary eye tracking applications have been and are currently being developed.. 2) FOR DATA ENTRY: Eye gazing technique employs Speech recognition and eye gaze tracking technology for use in Data entry tasks. It is efficient method while compared with other data entry methods (handwriting, mouse and keyboard and speech only) when carrying out the data entry task of tilling a form. Discussions regarding the relationships between efficiency, effectiveness, ergonomic quality, hedonic quality, naturalness, and familiarity and users preference are presented. The experimental results show that the majority of the users prefer using the proposed eye and speech system compared to the other form-filling methods even though such a method is neither the fastest nor the most accurate.

Nowadays, HUDs are becoming increasingly available in production cars (see fig. 6), and usually offer important data, like speedometer and navigation information. The information appears to hover above the engine hood, a few meters away from the driver's eyes.

4) TO OPERATE COMPUTER FOR PHYSICALLY DISABLED PERSONS.: The proposed application is meant to help to the people suffering from Motor Neuron Disease (MND), even though it can be useful for other disabilities that cause analogous communication impairments. Among the low technological solution graphical communication boards, which utilize pictures, drawings, abstracts symbols and/or text as symbols, can be very effective. Examples are communication books that are referenced by pointing, eye gaze, touching that is activated through switches or by indicating to a partner. Eye gaze in particular is typically used by children with severe motor disabilities.

3) FOR HEAD UP AND HEAD DOWN DIDPLAYS IN VEHICLES. The HUD makes it possible to project information directly into the driver's visual field. This principle is based on optical rules. An image is projected onto a glass window and is partially reflected. The reflected fraction is perceived by the observer as a virtual image with the distance of the image source.

5) CAPTURING MOVIE USING GAZE ALLIGNED CAMERA.

Head-fixed camera systems are widely known, but since they are aligned by the head only and not by the eyes they are not able to always look at what the cameraman is looking at, and the image quality is poor if no effort is made to stabilize the camera. Systems like Steady cam, in contrast, focus on image stabilization at the cost of restricting the cameraman in his actions. The prototype of a new mobile head-mounted camera system was developed that is continuously aligned with the orientation of gaze. In doing so, the biological gaze stabilization reflexes are used to keep the video camera stable on target. Applications like movie making through the eyes of an actor or documentary movies for sports and other activities are conceivable.

 It is a boon for persons suffering from Motor Neuron Disease since it does not require the movement of limbs.

VII. CONCLUSION

Figure 7. A gaze aligned head-mounted camera. The system consists of a pair of eye tracking video cameras that are laterally attached to swimming goggles and of a third camera above the forehead that moves with the eyes.

Here in this paper an effort is made to put light on the developments in the eye gaze technology. Few works have been dedicated to extracting user interest on the basis of gaze patterns. In other words, few works in the literature give a sense to the human-multimedia dialogue. The state of art lacks too on semantics associated to eye gaze/fixation communication between the human and multimedia documents. A wide variety of interdisciplinary eye tracking applications has been and is currently being developed. The state-of-the-art points to a scarcity of multimedia investigations. The domain of eye/gaze tracking in multimedia documents will lead to the explosion of multimedia usage in this area of research. It is hoped that it provides sufficient motivation to spur further interest in this fascinating domain.

VI.

ADVANTAGES

REFERENCES

It is one of the most versatile technologies under development. Since eye movement is the area under focus, it helps for easy computer interfacing. By processing the image of a wider field of view camera and determining the absolute eye position, the system quickly recovers from tracking errors due to sudden head movements or slow blinking. A data entry method that posses a balance of accuracy. Speed Naturalness Ease of use Hedonic quality compared to a method that is the fastest, Most accurate, Easy to use and that they are familiar with. HUDs minimize the mental workload for the driver and keep the increasing amount of information easily accessible.

1) [Kim99] Kim, K., and Ramakrishna, R.S. VisionBased Eye-Gaze Tracking for Human Computer Interface, Systems, Man and Cybernetics, IEEE SMC99 Conference Proc., No.2, pp.324-329, 1999. 2) BAOSHEN Hu AND MINGHUA QIU, A New Method for Human-Computer Interaction by using Eye Gaze, Proceedings of IEEE International Conference on Systems, Man and Cybernetics, October, 1994.