Article
Computer Science, Artificial Intelligence
Primesh Pathirana, Shashimal Senarath, Dulani Meedeniya, Sampath Jayarathna
Summary: This paper presents a comprehensive survey of single-user and multi-user gaze estimation approaches with deep learning. State-of-the-art methods are analyzed, and the limitations, challenges, and future directions of multi-user gaze estimation are discussed.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Mindaugas Vasiljevas, Robertas Damasevicius, Rytis Maskeliunas
Summary: Eye gaze interfaces enable users to control graphical user interfaces simply by looking at them, but using these interfaces can be demanding and lead to fatigue. To address these challenges, the authors propose a model based on biofeedback that allows for effective and sustainable control of computer interfaces using physiological signals. Experimental findings show that the proposed model effectively describes and explains performance dynamics during gaze control tasks, including subject variability, fatigue, and recovery.
Article
Chemistry, Analytical
Seungbong Lee, Jaehoon Jeong, Nahyun Kim, Manjae Shin, Sungmin Kim
Summary: Eye-gaze direction-tracking technology is important in various fields, and this study proposes a method to improve its speed and precision. By utilizing the shape of the pupil, the proposed method can estimate the rotational center point of the eye faster while maintaining accuracy. Experimental results demonstrate the effectiveness of this method in achieving higher speed and comparable precision to existing methods.
Article
Psychology, Mathematical
Niilo V. Valtakari, Roy S. Hessels, Diederick C. Niehorster, Charlotte Viktorsson, Par Nystroem, Terje Falck-Ytter, Chantal Kemner, Ignace T. C. Hooge
Summary: Computer-vision-based gaze estimation techniques directly estimate gaze direction from video recordings without the need for an eye tracker. This study aimed to identify usable methods for average researchers in fields like psychology or education and evaluated two toolkits, OpenFace and OpenGaze. OpenGaze was found to be accurate and precise enough for screen-based experiments, while OpenFace showed potential in sparse environments. However, OpenFace should not be used for drawing conclusions on measures like dwell duration.
BEHAVIOR RESEARCH METHODS
(2023)
Article
Psychology, Multidisciplinary
David Souto, Olivia Marsh, Claire Hutchinson, Simon Judge, Kevin B. Paterson
Summary: The development of gaze-controlled computer interfaces has been witnessed in the past twenty years for augmentative communication and assistive technology applications. Executive control is crucial for learning to use gaze-control, affecting technology uptake, with learning-induced plasticity contributing to improved performance of gaze-typing.
COMPUTERS IN HUMAN BEHAVIOR
(2021)
Article
Chemistry, Analytical
Michael Barz, Daniel Sonntag
Summary: The research focuses on automatically detecting visual attention to AOIs using pre-trained deep learning models for image classification and object detection. An evaluation framework based on the VISUS dataset and well-known performance metrics from the field of activity recognition was developed. The methods were systematically evaluated within this framework, discussing potentials and limitations, and proposing ways to improve the performance of future automatic visual attention detection methods.
Article
Education & Educational Research
Gianluca Schiavo, Nadia Mana, Ornella Mich, Massimo Zancanaro, Remo Job
Summary: The paper introduces an assistive reading tool that combines read-aloud technology with eye tracking to help struggling readers improve their reading comprehension. Results from a controlled experiment show that children using this tool significantly increased their comprehension scores, especially those with more inaccurate reading.
JOURNAL OF COMPUTER ASSISTED LEARNING
(2021)
Article
Acoustics
Ko-Feng Lee, Yen-Lin Chen, Chao-Wei Yu, Cheng-Lung Jen, Kai-Yi Chin, Chen-Wei Hung, Chih-Bo Wen
Summary: This study utilized a head-mounted camera to track eye behaviors and developed a low-cost efficient eye tracking system with a multi-point calibration algorithm. The system demonstrated high accuracy, precision, and speed, achieving stable detection results under different lighting sources.
JOURNAL OF LOW FREQUENCY NOISE VIBRATION AND ACTIVE CONTROL
(2021)
Article
Automation & Control Systems
Seunghyeb Ban, Yoon Jae Lee, Ki Jun Yu, Jae Won Chang, Jong-Hoon Kim, Woon-Hong Yeo
Summary: This study introduces a two-camera eye-tracking system and a data classification method for human-machine interfaces (HMIs). The system uses machine-learning technology to accurately control a robotic arm by continuously classifying gaze and eye directions in real time. The results show that the deep-learning classification algorithm achieves exceptional accuracy (99.99%) with a high number of actions per command (>= 64), outperforming other HMI systems.
ADVANCED INTELLIGENT SYSTEMS
(2023)
Article
Engineering, Biomedical
C. Leblond-Menard, S. Achiche
Summary: Weiming Shen is a professor at Huazhong University of Science and Technology (HUST) and his research interests include agent-based collaboration technology and applications, the Internet of Things, and big data analytics. He has published several books and over 500 papers, with over 14000 citations and an H-index of 56. Shen is a member of the American Society of Mechanical Engineers (ASME) and the Association for Computing Machinery (ACM).
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2023)
Article
Chemistry, Analytical
Moayad Mokatren, Tsvi Kuflik, Ilan Shimshoni
Summary: In this paper, a framework for 3D gaze estimation using a corneal imaging system is presented. The framework utilizes a headset with three cameras - a scene camera, an IR camera, and an RGB camera. Deep learning algorithms are used to detect pupil in IR and RGB images, creating a real-time 3D model of the eye. By calibrating the eye tracker transparently, the framework achieves low 3D gaze error and high accuracy in acquiring corneal images. The framework shows promising results in realistic settings and can be used in various mobile scenarios.
Article
Computer Science, Information Systems
Dean J. Lawrence, Hannah G. Imboden, Hussein K. Chebli, Maya S. Kabbash, Adnan Shaout
Summary: EyePaint is a system designed to assist individuals with physical disabilities in painting on canvas, using webcam-based gaze estimation and a custom graphical user interface. The system allows users to draw shapes on a virtual canvas before committing them to a physical canvas through low-accuracy gaze estimation.
JORDAN JOURNAL OF ELECTRICAL ENGINEERING
(2021)
Article
Engineering, Electrical & Electronic
Rawdha Karmi, Ines Rahmany, Nawres Khlifa
Summary: This article proposes a new method for gaze estimation that utilizes convolutional neural networks for head position estimation, Viola Jones' algorithm for eye area detection, and three different models for gaze estimation. The results from the validation experiments using the Columbia gaze database show that the proposed method achieves significantly improved accuracy compared to existing approaches.
SIGNAL IMAGE AND VIDEO PROCESSING
(2023)
Article
Robotics
Benjamin A. Newman, Reuben M. Aronson, Siddhartha S. Srinivasa, Kris Kitani, Henny Admoni
Summary: The HARMONIC dataset presents a large multimodal dataset capturing human interactions with a robotic arm in a shared autonomy setting designed for assistive eating tasks. It includes data views of various aspects such as eye gaze, body pose, and hand pose, providing valuable resources for researchers interested in intention prediction, human mental state modeling, and shared autonomy studies. The dataset offers data streams in formats like video and human-readable CSV and YAML files.
INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH
(2022)
Article
Psychology, Multidisciplinary
Inka Schmitz, Wolfgang Einhaeuser
Summary: This study investigates how people estimate gaze direction in screen-based communication and found that estimates are more accurate in the horizontal direction, biased towards the sender's head position, and influenced by the repetition of the same sender.
COMPUTERS IN HUMAN BEHAVIOR
(2023)
Article
Computer Science, Artificial Intelligence
Hakan Cevikalp, Burak Benligiray, Omer Nezih Gerek
PATTERN RECOGNITION
(2020)
Article
Computer Science, Artificial Intelligence
Llukman Cerkezi, Cihan Topal
PATTERN RECOGNITION
(2020)
Article
Biology
Selcan Kaplan Berkaya, Ilknur Ak Sivrikoz, Serkan Gunal
COMPUTERS IN BIOLOGY AND MEDICINE
(2020)
Article
Computer Science, Software Engineering
Atakan Dogan, Kemal Ebcioglu
Summary: The study introduces a parallel hardware hypervisor system to address power, performance, and scalability issues in cloud computing. This system, implemented entirely in special-purpose hardware, enables virtual supercomputers to share FPGA and ASIC resources in a cloud environment. Verification studies based on Verilog simulation have verified the functionality of the proposed system.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2022)
Article
Ecology
Selcan Kaplan Berkaya, Efnan Sora Gunal, Serkan Gunal
Summary: Deep learning-based image classification models are proposed for beehive monitoring, capable of recognizing different conditions and abnormalities with an accuracy of up to 99.07%, making them good candidates for smart beekeeping and beehive monitoring.
ECOLOGICAL INFORMATICS
(2021)
Article
Computer Science, Hardware & Architecture
Muhammet Yasin Pak, Serkan Gunal
Summary: This paper proposes a rule mining model based on sequential patterns for cross-domain opinion target extraction from product reviews in unknown domains. Experimental results show that the proposed model can extract opinion targets more accurately than previous studies.
COMPUTER SYSTEMS SCIENCE AND ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Cuneyt Akinlar, Hatice Kubra Kucukkartal, Cihan Topal
Summary: The paper proposes adding ellipse fit error as a shape prior regularization term to FCNs for image semantic segmentation, achieving promising results in pupil detection through experiments.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Burak Batmaz, Atakan Dogan
Summary: This article presents a hardware accelerator for CoAP server network stack, which shows significant advantages in performance, latency, and power consumption compared to the software implementation. It can be a powerful solution for addressing the performance and power consumption issues of resource-constrained IoT devices.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Haydar Kilic, Salim Ceyhan, Omer Nezih Gerek
Summary: This article introduces new anisotropic image filters and compares their performances with existing isotropic and anisotropic filters. The developed filters are evaluated based on their image noise removal performance and the edge preserving properties of the filtered images. The findings suggest that the new metrics perform well against other anisotropic metrics and preserve edges better than other filters, making them a plausible noise removal tool in image processing.
DIGITAL SIGNAL PROCESSING
(2022)
Article
Chemistry, Multidisciplinary
Mohamed Lemine Sidi, Serkan Gunal
Summary: This paper proposes a Purely Entity-based Semantic Search Approach for Information Retrieval (PESS4IR) to improve document retrieval. The approach includes its own entity linking and inverted indexing methods, as well as an appropriate ranking method. The experiments show that the approach achieves good performance on queries with rich annotations.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Artificial Intelligence
Huseyin Gunduz, Cuneyd Nadir Solak, Serkan Gunal
Summary: In this study, a new and effective model for the automatic segmentation of diatoms based on image processing and deep learning algorithms is proposed. Through extensive experimental work, the performance of the proposed segmentation model is measured and verified to surpass previous works.
TURKISH JOURNAL OF ELECTRICAL ENGINEERING AND COMPUTER SCIENCES
(2022)
Article
Multidisciplinary Sciences
Mehmet Koc, Semih Ergin, Mehmet Bilginer Gulmezoglu, Mehmet Fidan, Omer Nezih Gerek, Atalay Barkana
Summary: This study investigates the classification performances and characteristics of two variations of the common vector approach (CVA) for images with shared backgrounds, including binary images. It finds that the discriminative CVA method carries certain risks when the dimension of the feature vector is lower than the number of training samples. Additionally, it observes that CVA outperforms its discriminative version in the classification of binary images. The study highlights the importance of considering the training data size when applying CVA methods.
ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
(2022)
Proceedings Paper
Engineering, Electrical & Electronic
Mukunthan Tharmakulasingam, Cihan Topal, Anil Fernando, Roberto La Ragione
2020 IEEE INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS (ICCE)
(2020)
Proceedings Paper
Engineering, Biomedical
Mukunthan Tharmakulasingam, Cihan Topal, Anil Fernando, Roberto La Ragione
ICBBE 2019: 2019 6TH INTERNATIONAL CONFERENCE ON BIOMEDICAL AND BIOINFORMATICS ENGINEERING
(2019)
Proceedings Paper
Engineering, Electrical & Electronic
Mehmet Senses, Cihan Topal
2019 27TH SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE (SIU)
(2019)
