• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.2
• /
• pp.477-483
• /
• 2004

This paper describes preliminary results we have obtained in developing a computer vision system based on active IR illumination for real time gaze tracking for interactive graphic display. Unlike most of the existing gaze tracking techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person our gaze tracker can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using the Generalized Regression Neural Networks(GRNN). With GRNN, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. Futhermore, the mapping function can generalize to other individuals not used in the training. The effectiveness of our gaze tracker is demonstrated by preliminary experiments that involve gaze-contingent interactive graphic display.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.8
• /
• pp.43-51
• /
• 2024

This paper proposes a method for acquiring and analyzing ocular data using a special-purpose diver mask targeted at divers who primarily engage in underwater activities. This involves tracking the user's gaze with the help of a custom-built ocular dataset and a YOLOv8-nano model developed for this purpose. The model achieved an average processing time of 45.52ms per frame and successfully recognized states of eyes being open or closed with 99% accuracy. Based on the analysis of the ocular data, a gaze tracking algorithm was developed that can map to real-world coordinates. The validation of this algorithm showed an average error rate of about 1% on the x-axis and about 6% on the y-axis.

• Journal of Korea Multimedia Society
• /
• v.18 no.10
• /
• pp.1251-1260
• /
• 2015

The measurement and evaluation of an audience's visual attention are crucial processes for the scientific production of various contents and services. It has been alleged that a vector exists between objects on a screen, but this has not been explored empirically. This study tested whether or not there are any positive correlations between screen vectors and viewers' eye movements. Participants were exposed to four groups of pictures representing an independent vector, continuing vector, converging vector, and diverging vector. The results showed that vectors on screens induced viewers' eye movements to the vector object by the directivity of the visual stimuli.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.5
• /
• pp.343-350
• /
• 2019

Map labels are the most recognizable map elements using the human visual system because they are essentially a natural language. In this study, an experiment was conducted using an eye-tracker to objectively record and analyze the response of subjects regarding visual attention to map labels. A primary building object was identified by analyzing visit counts, average visit duration, fixation counts, and the average fixation duration of a subject's gaze for an area of interest acquired using the eye-tracker. The unmarked rate of map labels in Google map, Naver map, and Daum map was calculated. As a result, this rate exceeded fifty-one percent, with the lowest rate recorded for Google map. It is expected that the results of this study will contribute to an increase in the diversity of research in terms of the spatial cognition approach for map labels, which is more helpful to users than the existing body of work on methods of expression for labels.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.522-526
• /
• 2019

추출 요약(Extractive summarization)은 문서내에 주요한 요약정보가 되는 문장 또는 단어를 추출하여 요약을 생성하는 기법이다. 딥러닝 기법들이 많이 발전하면서 요약 기법에도 sequence-to-sequence와 같은 많은 시도들이 있었지만 대부분의 방법론들은 딥러닝의 모델 구조관점으로 접근하거나 요약에 있어서 단순히 입력 텍스트를 넣고 알고리즘이 처리하는 머신 리딩(Machine reading)관점으로 접근한다. 텍스트 요약 태스크 자체는 사람이 텍스트에 대한 정보 파악을 요약문을 통해 빠르게 하고 싶은 궁극적인 목표가 있으므로, 사람이 텍스트 요약에 필요한 인지처리과정을 반영할 필요가 있다. 결국, 기존의 머신 리딩보다는 휴먼 리딩(Human reading)에 관한 이해와 구조적 접근이 필요하다. 따라서 본 연구는 휴먼 리딩을 위한 인지처리과정을 위해 아이트래킹 데이터 기반의 새로운 추출 요약 모델을 제안한다.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2008.10a
• /
• pp.947-950
• /
• 2008

This paper reports of traceable and repeatable time resolved colorimetric measurements of the luminance- and colour variation occurring during the frame refresh period of FPD's. During one frame period a number of measurements are performed, and for the dynamic artifacts a simulated smooth pursuit eye tracking algorithm is included.

• 한국항공운항학회:학술대회논문집
• /
• 2006.11a
• /
• pp.87-90
• /
• 2006

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2010.11a
• /
• pp.989-990
• /
• 2010

• Journal of the Ergonomics Society of Korea
• /
• v.34 no.5
• /
• pp.377-399
• /
• 2015

Objective: The aim of this study is to understand and identify the critical issues in vision research area using content analysis and network analysis. Background: Vision, the most influential factor in information processing, has been studied in a wide range of area. As studies on vision are dispersed across a broad area of research and the number of published researches is ever increasing, a bibliometric analysis towards literature would assist researchers in understanding and identifying critical issues in their research. Method: In this study, content and network analysis were applied on the meta-data of literatures collected using three search keywords: 'visual search', 'eye movement', and 'eye tracking'. Results: Content analysis focuses on extracting meaningful information from the text, deducting seven categories of research area; 'stimuli and task', 'condition', 'measures', 'participants', 'eye movement behavior', 'biological system', and 'cognitive process'. Network analysis extracts relational aspect of research areas, presenting characteristics of sub-groups identified by community detection algorithm. Conclusion: Using these methods, studies on vision were quantitatively analyzed and the results helped understand the overall relation between concepts and keywords. Application: The results of this study suggests that the use of content and network analysis helps identifying not only trends of specific research areas but also the relational aspects of each research issue while minimizing researchers' bias. Moreover, the investigated structural relationship would help identify the interrelated subjects from a macroscopic view.

• Journal of Convergence for Information Technology
• /
• v.11 no.11
• /
• pp.51-56
• /
• 2021

In this paper, performance degradation in the cross-eye jamming due to amplitude mismatch of two jamming antennas is considered. The mismatch of the amplitude ratio is modeled as a random variable with a normal distribution of the difference between the actual amplitude ratio and the nominal amplitude ratio due to mechanical defects. In the proposed analytic performance analysis, the first-order Taylor series expansion and the second-order Taylor series expansion is adopted. Performance measure of the cross-eye jamming is the mean square difference (MSD). The analytically derived MSD is validated by comparing the analytically derived MSD with the first-order Taylor series-based simulation-based MSD and the second-order Taylor series-based simulation-based MSD. It shows that the analysis-based MSD is superior to the Monte-Carlo-based MSD, which has a high calculation cost.