Binocular eye-gaze tracking can be used to estimate the point-of-gaze (POG) of a subject in real-world three-dimensional (3D) space using the vergence of the eyes. In this thesis, a novel non-contact, model-based technique for 3D POG estimation is presented. The non-contact system allows people to select real-world objects in 3D physical space using their eyes, without the need for head-mounted equipment. Using a model-based POG estimation algorithm allows for free head motion and a single stage of calibration. The users were free to naturally move and reorient their heads while operating the system, within an allowable headspace of 3.2 x 9.2 x 14 cm. A rela tively high precision, as measured by the standard deviation of the 3D POG estimates, was measured to be 0.26 cm and was achieved with the use of high speed sampling and digital filtering techniques. When observing points in a 3D volume, large head and eye rotations are far more common than when observing a 2D screen. A novel corneal reflection pattern matching algorithm is presented for increasing image feature tracking reliability in the presence of large eye rotations. It is shown that an average accuracy of 3.93 cm was achieved over seven different subjects and a workspace volume of 30 x 23 x 25 cm (width x height x depth). An example application is presented illustrating the use of the 3D POG as a human computer interface in a 3D game of Tic-Tac-Toe on a 3 x 3 x 3 volumetric display.

Binocular eye-gaze tracking can be used to estimate the point-of-gaze (POG) of a subject in real-world three-dimensional (3D) space using the vergence of the eyes. In this thesis, a novel non-contact, model-based technique for 3D POG estimation is presented. The non-contact system allows people to select real-world objects in 3D physical space using their eyes, without the need for head-mounted equipment. Using a model-based POG estimation algorithm allows for free head motion and a single stage of calibration. The users were free to naturally move and reorient their heads while operating the system, within an allowable headspace of 3.2 x 9.2 x 14 cm. A rela tively high precision, as measured by the standard deviation of the 3D POG estimates, was measured to be 0.26 cm and was achieved with the use of high speed sampling and digital filtering techniques. When observing points in a 3D volume, large head and eye rotations are far more common than when observing a 2D screen. A novel corneal reflection pattern matching algorithm is presented for increasing image feature tracking reliability in the presence of large eye rotations. It is shown that an average accuracy of 3.93 cm was achieved over seven different subjects and a workspace volume of 30 x 23 x 25 cm (width x height x depth). An example application is presented illustrating the use of the 3D POG as a human computer interface in a 3D game of Tic-Tac-Toe on a 3 x 3 x 3 volumetric display.

Binocular eye-gaze tracking can be used to estimate the point-of-gaze (POG) of a subject in real-world three-dimensional (3D) space using the vergence of the eyes. In this thesis, a novel non-contact, model-based technique for 3D POG estimation is presented. The non-contact system allows people to select real-world objects in 3D physical space using their eyes, without the need for head-mounted equipment. Using a model-based POG estimation algorithm allows for free head motion and a single stage of calibration. The users were free to naturally move and reorient their heads while operating the system, within an allowable headspace of 3.2 x 9.2 x 14 cm. A rela tively high precision, as measured by the standard deviation of the 3D POG estimates, was measured to be 0.26 cm and was achieved with the use of high speed sampling and digital filtering techniques. When observing points in a 3D volume, large head and eye rotations are far more common than when observing a 2D screen. A novel corneal reflection pattern matching algorithm is presented for increasing image feature tracking reliability in the presence of large eye rotations. It is shown that an average accuracy of 3.93 cm was achieved over seven different subjects and a workspace volume of 30 x 23 x 25 cm (width x height x depth). An example application is presented illustrating the use of the 3D POG as a human computer interface in a 3D game of Tic-Tac-Toe on a 3 x 3 x 3 volumetric display. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

An objective methodology to assess visual acuity (VA) in infants was developed. The methodology is based on the analysis of visual scanning parameters when visual stimuli consisting of homogeneous targets and a target with gratings (TG) are presented. The percentage of time on the TG best predicted the ability of the subject to discriminate between the targets. Using this parameter, the likelihood ratio test was used to test the hypothesis that the TG was discriminated. VA is estimated as the highest spatial frequency for which the probability of false positive is lower than the probability of false negative for stimuli with lower spatial frequencies. VA estimates of 9 adults had an average error of 0.06 logMAR with a testing time of 3.5 minutes. These results suggest that if the attention of infants can be consistently maintained the new methodology will enable more accurate assessment of VA in infants.

An objective methodology to assess visual acuity (VA) in infants was developed. The methodology is based on the analysis of visual scanning parameters when visual stimuli consisting of homogeneous targets and a target with gratings (TG) are presented. The percentage of time on the TG best predicted the ability of the subject to discriminate between the targets. Using this parameter, the likelihood ratio test was used to test the hypothesis that the TG was discriminated. VA is estimated as the highest spatial frequency for which the probability of false positive is lower than the probability of false negative for stimuli with lower spatial frequencies. VA estimates of 9 adults had an average error of 0.06 logMAR with a testing time of 3.5 minutes. These results suggest that if the attention of infants can be consistently maintained the new methodology will enable more accurate assessment of VA in infants.

This thesis proposes a novel approach for efficiently estimating gaze points in dual camera transportation datasets. Traditional methods for gaze point estimation are dependent on large amounts of labeled data, which can be both expensive and time-consuming to collect. Additionally, alignment and calibration of the two camera views present significant challenges. To overcome these limitations, this thesis investigates the use of self-learning techniques such as semi-supervised learning and self-training, which can reduce the need for labeled data while maintaining high accuracy. The proposed method is evaluated on the DGAZE dataset and achieves a 57.2\% improvement in performance compared to the previous methods. This approach can prove to be a valuable tool for studying visual attention in transportation research, leading to more cost-effective and efficient research in this field. / Master of Science / This thesis presents a new method for efficiently estimating the gaze point of drivers while driving, which is crucial for understanding driver behavior and improving transportation safety. Traditional methods require a lot of labeled data, which can be time-consuming and expensive to obtain. This thesis proposes a self-learning approach that can learn from both labeled and unlabeled data, reducing the need for labeled data while maintaining high accuracy. By training the model on labeled data and using its own estimations on unlabeled data to improve its performance, the proposed approach can adapt to new scenarios and improve its accuracy over time. The proposed method is evaluated on the DGAZE dataset and achieves a 57.2\% improvement in performance compared to the previous methods. Overall, this approach offers a more efficient and cost-effective solution that can potentially help improve transportation safety by providing a better understanding of driver behavior. This approach can prove to be a valuable tool for studying visual attention in transportation research, leading to more cost-effective and efficient research in this field.

This thesis presents a novel system that estimates the point-of-gaze (where a person is looking at) remotely while allowing for free head movements and minimizing personal calibration requirements. The point-of-gaze is estimated from the pupil and corneal reflections (virtual images of infrared light sources that are formed by reflection on the front corneal surface, which acts as a convex mirror) extracted from eye images captured by video cameras. Based on the laws of geometrical optics, a detailed general mathematical model for point-of-gaze estimation using the pupil and corneal reflections is developed. Using this model, the full range of possible system configurations (from one camera and one light source to multiple cameras and light sources) is analyzed. This analysis shows that two cameras and two light sources is the simplest system configuration that can be used to reconstruct the optic axis of the eye in 3-D space, and therefore measure eye movements, without the need for personal calibration. To estimate the point-of-gaze, a simple single-point personal calibration procedure is needed. The performance of the point-of-gaze estimation depends on the geometrical arrangement of the cameras and light sources and the method used to reconstruct the optic axis of the eye. Using a comprehensive simulation framework developed from the mathematical model, the performance of several gaze estimation methods of varied complexity is investigated for different geometrical system setups in the presence of noise in the extracted eye features, deviation of the corneal shape from the ideal spherical shape and errors in system parameters. The results of this investigation indicate the method(s) and geometrical setup(s) that are optimal for different sets of conditions, thereby providing guidelines for system implementation. Experimental results with adults, obtained with a system that follows those guidelines, exhibit RMS point-of-gaze estimation errors of 0.4-0.6º of visual angle (comparable to the best commercially available systems, which require multiple-point personal calibration procedures). Preliminary results with infants demonstrate the ability of the proposed system to record infants' visual scanning patterns, enabling applications that are very difficult or impossible to carry out with previously existing technologies (e.g., study of infants' visual and oculomotor systems).

This thesis presents a novel system that estimates the point-of-gaze (where a person is looking at) remotely while allowing for free head movements and minimizing personal calibration requirements. The point-of-gaze is estimated from the pupil and corneal reflections (virtual images of infrared light sources that are formed by reflection on the front corneal surface, which acts as a convex mirror) extracted from eye images captured by video cameras. Based on the laws of geometrical optics, a detailed general mathematical model for point-of-gaze estimation using the pupil and corneal reflections is developed. Using this model, the full range of possible system configurations (from one camera and one light source to multiple cameras and light sources) is analyzed. This analysis shows that two cameras and two light sources is the simplest system configuration that can be used to reconstruct the optic axis of the eye in 3-D space, and therefore measure eye movements, without the need for personal calibration. To estimate the point-of-gaze, a simple single-point personal calibration procedure is needed. The performance of the point-of-gaze estimation depends on the geometrical arrangement of the cameras and light sources and the method used to reconstruct the optic axis of the eye. Using a comprehensive simulation framework developed from the mathematical model, the performance of several gaze estimation methods of varied complexity is investigated for different geometrical system setups in the presence of noise in the extracted eye features, deviation of the corneal shape from the ideal spherical shape and errors in system parameters. The results of this investigation indicate the method(s) and geometrical setup(s) that are optimal for different sets of conditions, thereby providing guidelines for system implementation. Experimental results with adults, obtained with a system that follows those guidelines, exhibit RMS point-of-gaze estimation errors of 0.4-0.6º of visual angle (comparable to the best commercially available systems, which require multiple-point personal calibration procedures). Preliminary results with infants demonstrate the ability of the proposed system to record infants' visual scanning patterns, enabling applications that are very difficult or impossible to carry out with previously existing technologies (e.g., study of infants' visual and oculomotor systems).