Home Position Recognition Methods Using Polarizing Films for an

Eye-gaze Interface System

Kohichi Ogata

, Kensuke Sakamoto

and Shingo Niino

Graduate School of Science and Technology, Kumamoto University,

2-39-1 Kurokami, Chuo-ku, Kumamoto 860-8555, Japan

Faculty of Engineering, Kumamoto University, 2-39-1 Kurokami, Chuo-ku, Kumamoto 860-8555, Japan

Keywords:

Polarizing Films, Eye, Gaze, Interface, Head Movement.

Abstract:

In eye-gaze interface systems, users’ head movements during use result in detection errors. This problem

causes inaccurate positioning of the mouse pointer on the display screen. A mechanism that allows the user to

recognize the home position, which is an appropriate position for the head while using an eye-gaze interface

system, can provide a useful solution to the problem because the user can then simply adjust his/her head

position. The implementation of such a mechanism does not require special equipment, such as a position

sensor to detect head movement and calculate compensation. We thus propose in this paper methods for

recognizing the home position using polarizing ﬁlms. Taking advantage of the characteristics of polarizing

ﬁlms, we propose two guidance methods to help an eye-gaze interface system user recognize whether or

not his/her head is in the home position and adjust the position of the head if needed. The results of our

experiments reveal that our proposed methods improve the usability of eye-gaze interface systems, and that

one of our methods is more effective than the other. Therefore, our mechanism is useful in producing simple

and low-cost interface systems.

1 INTRODUCTION

Users’ head movements cause detection errors in eye-

gaze interface systems. This leads to the undesir-

able positioning of the mouse pointer on the display

screen. Therefore, it is necessary to somehow com-

pensate for the troublesome movements in order to

reduce errors.

A solution to the problem is to monitor head-

related movements with multiple cameras (Talmi and

Liu, 1999; Yoo and Chung, 2005; Noureddin et al.,

2005). Such a method requires additional equipment

to monitor movements and calculate the necessary

compensationdue to the movementbased on the mon-

itored deviation, which results in complex and costly

systems. However, if a user can appropriately adjust

the position of his/her head before pointing or click-

ing through an eye-gaze interface, the pointing error

can be reduced. Therefore, taking into account the

mechanism by which the user can recognize the de-

viation of the head from the home position, which is

an appropriate position for the head while an eye-gaze

interface is being used, is useful for creating simple,

but effective and low-cost, interface systems. In order

to realize such mechanisms, we propose methods for

recognizing the head position using polarizing ﬁlms

(Ogata et al., 2012). Taking advantage of the charac-

teristics of polarizing ﬁlms, we propose two guidance

methods to help recognize whether or not the user’s

head is in the home position and to accordingly adjust

it. Experiments are conducted and a questionnaire is

distributed in order to evaluate the usefulness of our

proposed methods. The results contribute to provid-

ing users with a useful reference in order to recognize

the home position while using eye-gaze interface sys-

tems. Because they render unnecessary the detection

of the position of the head, our guidance methods en-

able the creation of simple and cheap eye-gaze inter-

face systems.

2 EYE-GAZE INTERFACE

SYSTEM

Our eye-gaze interface system (Yonezawa et al.,

2008; Yonezawa et al., 2009; Yonezawa et al., 2010;

Yonezawa, 2010) consists of a small visible light

275

Ogata K., Sakamoto K. and Niino S..

Home Position Recognition Methods Using Polarizing Films for an Eye-gaze Interface System.

DOI: 10.5220/0005048902750280

In Proceedings of the 11th International Conference on Signal Processing and Multimedia Applications (SIGMAP-2014), pages 275-280

ISBN: 978-989-758-046-8

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

camera (Kyohritsu JPP-CM25F 1/3 inch CMOS 0.25

Mpixel) and a desktop computer (CPU: Intel Core i5-

650 3.20 GHz, Memory: 2.92 GB, OS: Windows XP)

with an image-capture board (Imagination PXC200).

The system is capable of real-time processing with

320 × 240 pixels at 30 fps. The visible light cam-

era is attached to the user’s goggles to capture images

of the right eye, and a computer display and a 20 W

ﬂuorescent table lamp are located in front of the user.

We use a chin support in our system to provide a rest

position and reduce the user’s head movements. How-

ever, the development of a home position recognition

method can render the chin support unnecessary for

our system, and can hence expand its range of appli-

cation.

2.1 Semi-dynamic Calibration (Ogata

and Matsumoto, 2012)

In an eye-gaze-driven mouse-pointing system, the

user perceives degradation in pointing accuracy

through a discrepancy between the true eye-gaze and

the position of the mouse pointer. Therefore, a semi-

dynamic calibration (Ogata and Matsumoto, 2012)

is installed in the system to update a mapping func-

tion between the center of the iris on the captured

eye image and its corresponding calculated eye-gaze

point on the display screen. The user can activate the

semi-dynamic calibration mode anytime during use

by blinking his/her eye for one second or longer.

3 PROPOSED METHODS USING

POLARIZING FILMS

Figure 1 shows the schematic characteristics of a po-

larizing ﬁlm. As is well known, the ﬁlm can only be

transmitted by light whose polarizing axis is parallel

to the transmission axis of the polarizing ﬁlm. If two

ﬁlms are used and their transmission axes are perpen-

dicular to each other, the light can turn off. If we can

apply these characteristics to a method that informs a

user of his or her posture, it can serve as a low-cost

but useful guidance tool. In this paper, we use paired

polarizing ﬁlms for the user to determine whether or

not his/her head is at the home position.

Figure 2 shows the arrangement of the polarizing

ﬁlms (C-TASK CO., LTD, 0.2 mm thick, transmit-

tance 38%, degree of polarization 95% or more). The

equipment for the two proposed methods Guide 1 and

Guide 2 is arranged around the computer display. The

details of each method are as follows.

In Guide 1, paired polarizing ﬁlms are placed be-

Figure 1: Schematic characteristics of a polarizing ﬁlm.

Figure 2: Arrangement of polarizing ﬁlms.

low the display. Four polarizing ﬁlms are arranged

so that their transmission axes follow the directions

shown in Figure 3(a). One set each of this mosaic-

like design of ﬁlm is attached to a white board and

a transparent plastic plate, as shown in Figure 3(b).

The white board and the plastic plate are arranged at a

distance, as shown in Figure 3(c). If the user’s head is

at the home position, he/she sees white or gray polar-

izing ﬁlms because of their parallel transmission axes.

If the head is not at the home position, black zones ap-

pear on the polarizing ﬁlms because the transmission

axis of the polarizing ﬁlm on the transparent plastic

plate is perpendicular to that on the white board in

these zones. An example of such situations is shown

in Figure 3(d), where horizontal black zones appear

because the head is out of position in the upper direc-

tion.

In Guide 2, we use light-emitting diodes (LEDs)

(Daiso Japan, D-011 LED LIGHT A no.5) and polar-

izing ﬁlms. As shown in Figure 4(a), a plastic bar

is attached to the front cover of the LED so that the

axis of the bar is perpendicular to the face of the front

cover. A polarizing ﬁlm is attached to the tip of the

bar so that the axis of the bar is perpendicular to the

ﬁlm. Another polarizing ﬁlm is attached to the front

of the user’s goggles, as shown in Figure 4(b). Be-

cause the transmission axes of the polarizing ﬁlms at-

SIGMAP2014-InternationalConferenceonSignalProcessingandMultimediaApplications

276

Figure 3: Arrangement of polarizing ﬁlms in Guide 1.

tached to the LED and the goggles are perpendicular

to each other, the light from the LED cannot reach the

user’s eyes if his/her head is at the home position. As

shown in Figure 2, three sets of polarizing ﬁlm with

LED are arranged on the frame of the display.

Figure 5 shows three examples of the appearance

of the LED light. In case the head is at the home po-

sition, the light from the LED cannot reach the user’s

eyes, as shown in (a). The other examples correspond

to cases where the user bends his or her neck and the

head consequently moves in the corresponding direc-

tion.

Figure 4: Arrangement of polarizing ﬁlms in Guide 2.

Figure 5: Examples of the LED lighting up.

4 EXPERIMENTS

Experiments to test our proposed methods were con-

ducted without the chin support. Five subjects with

normal eyesight participated in the experiments. A

17-inch liquid-crystaldisplay (LCD) with 1024 × 768

pixels was placed in front of the subject. The dis-

tance between the screen and the subject’s head var-

ied from 60 to 80 cm. The variation arose because the

chin support was not used, and thus each subject as-

sumed a head position according to his/her comfort.

Each subject was asked to gaze at one of 25 target

HomePositionRecognitionMethodsUsingPolarizingFilmsforanEye-gazeInterfaceSystem

277

points that randomly appeared one after another on

the screen. The experiments consisted of three stages,

involving the use of the eye-gaze interface (a) with-

out the guides, (b) with the recognition method Guide

1, and (c) with the recognition method Guide 2. The

subjects were randomly ordered in order to avoid the

order effect in these stages. Semi-dynamic calibration

was used if necessary.

5 EXPERIMENTAL RESULTS

Table 1 shows the pointing accuracy for all subjects

in the use of the eye-gaze interface. Each value de-

notes the average accuracy over the 25 target points.

The pointing accuracy for each target point was calcu-

lated from image data of 30 frames that correspond to

1 s. The accuracy varies with the users and the meth-

ods. The results suggest that the pointing accuracy

of the eye-gaze system using our recognition meth-

ods is comparable to that without the methods. We

think that because semi-dynamic calibration worked

effectively to update the mapping function, the three

conditions had no signiﬁcant effect on improving the

degree of accuracy.

Table 2 shows the time required for each trial.

As shown in the table, the average time required for

Guide 2 is 270.8 s, in contrast to 427.4 s without the

guides. The number of times the semi-dynamic cal-

ibration was activated in each trial is also shown in

Table 3. The shorter times needed suggest that the

guide systems are useful in recognizing the position

of the user’s head and reducing the frequency of situ-

ations where the mouse pointer on the screen does not

follow the user’s gaze.

We performed usability evaluation at the end of

the experiment for each subject. Table 4 shows the

questionnaire items for evaluation. The subjects were

asked questions related to ease of use, accuracy, and

comfort of the system for each of the three conditions:

Without Guides, Guide 1, and Guide 2. Items regard-

ing the recognition of the home position and appropri-

ate adjustment of the head position were added to the

questionnaires for Guide 1 and Guide 2. Because a

polarizing ﬁlm is attached to the front of the goggles,

as shown in Figure 4(b), an item regarding visibil-

ity was added to the questionnaire for Guide 2. The

ﬁnal item was the subjects’ overall impression of the

usefulness of the guides.

Table 5 shows the results of the questionnaire.

Items (1) to (14) were evaluated on a grading scale

ranging from 1 (poor) to 5 (good). Item (15) was

evaluated according to the number of subjects who

preferred the method. As indicated in item (15), all

Table 1: Results of pointing accuracy.

Accuracy [pixels]

Subjects Without Guides Guide 1 Guide 2

Subject 1 34.6 44.5 44.1

Subject 2 57.2 65.3 56.3

Subject 3 47.8 36.5 26.1

Subject 4 61.3 84.7 80.7

Subject 5 75.5 53.6 55.3

Average 55.3 56.9 52.5

Table 2: Time required in each trial.

Time [s]

Subjects Without Guides Guide 1 Guide 2

Subject 1 374.9 248.6 189.4

Subject 2 409.1 570.3 307.9

Subject 3 674.0 161.1 258.3

Subject 4 312.1 266.1 252.1

Subject 5 366.9 226.3 346.3

Average 427.4 294.5 270.8

Table 3: The number of uses of semi-dynamic calibration

in each trial.

Number [times]

Subjects Without Guides Guide 1 Guide 2

Subject 1 11 5 3

Subject 2 5 7 6

Subject 3 21 1 4

Subject 4 8 7 5

Subject 5 6 3 3

Average 10.2 4.6 4.2

subjects preferred the system with guides. Among

the three conditions, Guides 1 and 2 recorded higher

scores on the three common items regarding ease of

use, accuracy, and comfort. A comparison of the

scores for items (7), (8), (12), and (13) shows that

Guide 2 is superior to Guide 1 in terms of recog-

nizing the home position and appropriately adjusting

the head position. In a comment form provided in

the questionnaire, a subject pointed out that Guide 2

was more suitable for shortsighted people than Guide

1. Another subject pointed out that the polarizing

ﬁlm attached to the goggles helped improve visibility.

This suggests that the ﬁlm eliminated part of the light

from the display and made it easier for the subject to

see.

6 CONCLUSIONS

In this paper, we proposed two methods for recog-

nizing the user’s head position using polarizing ﬁlms

SIGMAP2014-InternationalConferenceonSignalProcessingandMultimediaApplications

278

Table 4: Items in the questionnaire regarding ease of use.

Without Guides

(1) Easy to move the mouse pointer to a target.

(2) Accuracy improved after semi-dynamic

calibration.

(3) Did not feel tired after pointing 25 targets.

Guide 1

(4) Easy to move the mouse pointer to a target.

(5) Accuracy improved after semi-dynamic

calibration.

(6) Did not feel tired after pointing 25 targets.

(7) Easy to recognize the home position

through the guide.

(8) Easy to move the head to the home posi-

tion.

Guide 2

(9) Easy to move the mouse pointer to a target.

(10) Accuracy improved after semi-dynamic

calibration.

(11) Did not feel tired after pointing 25 targets.

(12) Easy to recognize the home position

through the guide.

(13) Easy to move the head to the home posi-

tion.

(14) The polarizing ﬁlm did not interfere with

visibility.

Overall impression

(15) Which is easier, with or without the

guides?

in an eye-gaze interface system. Our experiments

showed that the pointing accuracy of our proposed

methods is comparable to that of the conventional

method because semi-dynamic calibration works ef-

fectively to update the mapping function between the

center of the iris in the captured eye image and the

calculated eye-gaze point on the display screen. How-

ever, the guides are useful in reducing the frequency

of activation of the semi-dynamic calibration method

because the user can easily recognize the home po-

sition through the guides. Therefore, our proposed

methods help reduce the time required to point accu-

rately on the display in eye-gaze interface systems.

The results from our experiments and the question-

naire revealed that the proposed methods improve the

usability of our system.

Although the user in our method needs to move

his/her head to the home position before pointing or

clicking in an eye-gaze interface, our proposed meth-

ods are a reasonable solution as shown in the ex-

periments. They permit free head movement except

when pointing or clicking. Because our eye-gaze in-

terface system captures eye images through a cam-

era attached to goggles, additional equipment will be

Table 5: Results of the questionnaire with grading scale

ranging from 1 (poor) to 5 (good). Item (15) indicates the

number of the subjects who chose between using the system

with and without the guides.

Without Guides

(1) 2.6

(2) 3.4

(3) 2.8

Guide 1

(4) 3.4

(5) 4.4

(6) 3.0

(7) 3.2

(8) 3.0

Guide 2

(9) 3.6

(10) 4.4

(11) 3.4

(12) 4.6

(13) 4.2

(14) 3.6

Overall impression

(15) with: 5, without: 0

required if we try to introduce a compensation tech-

nique, such as face detection-based approaches. Our

methods proposed here is effective and avoids over-

complicating the system. Moreover, because it is not

necessary to modify the internal mechanism of the

eye-gaze system in order to implement our methods,

they serve as a versatile assistance tool. Further im-

provement is needed to develop more sophisticated

guidance tools.

ACKNOWLEDGEMENTS

Part of this work was supported by Grant-in-Aid for

the Japan Society for the Promotion of Science.

REFERENCES

Noureddin, B., Lawrence, P., and Man, C. (2005). A non-

contact device for tracking gaze in a human computer

interface. Comput. Vis. Image Und., 98(1):52–82.

Ogata, K. and Matsumoto, K. (2012). Semi-dynamic cali-

bration for eye gaze pointing system based on image

processing. In Proceedings of International Confer-

ence on Signal Processing and Multimedia Applica-

tions (SIGMAP 2012), pages 233–236. SciTePress.

Ogata, K., Sakamoto, K., and Niino, S. (2012). A study

on head position recognition methods for eye gaze in-

terface system. In Proceedings of 11th Forum on In-

HomePositionRecognitionMethodsUsingPolarizingFilmsforanEye-gazeInterfaceSystem

279

formation Technology (FIT2012), pages 497–498. (in

Japanese).

Talmi, K. and Liu, J. (1999). Eye and gaze tracking for visu-

ally controlled interactive stereoscopic displays. Sig-

nal Processing: Image Communication, 14:799–810.

Yonezawa, T. (2010). A Study on Development of Eye-gaze

Input System Using Iris Center Detection Based on

Image Processing. PhD thesis, Kumamoto University.

(in Japanese).

Yonezawa, T., Ogata, K., Matsumoto, K., Hirase, S., Shi-

ratani, K., Kido, D., and Nishimura, M. (2010). A

study on reducing the detection errors of the center of

the iris for an eye-gaze interface system. IEEJ Trans.

EIS, 130-C(3):442–449. (in Japanese).

Yonezawa, T., Ogata, K., and Shiratani, K. (2008). An

algorithm for detecting the center of the iris using

a color image and improvements of its processing

speed. IEEJ Trans. EIS, 128-C(2):253–259. (in

Japanese).

Yonezawa, T., Ogata, K., and Shiratani, K. (2009). An

algorithm for detecting the center of the iris using

a color image and improvements of its processing

speed. Electronics and Communications in Japan,

92(4):1–8. (Translated paper).

Yoo, D. H. and Chung, M. J. (2005). A novel non-intrusive

eye gaze estimation using cross-ratio under large head

motion. Computer Vision and Image Understanding,

98:25–51.

SIGMAP2014-InternationalConferenceonSignalProcessingandMultimediaApplications

280