![]() The viewed plane in remote eye trackers (RET) usually is a computer monitor, and in head-mounted eye trackers (HMET) usually is an image from a scene camera to represent the user’s field-of-view. In general, video-based eye-tracking methods extract features from the eye image (e.g., pupil center, iris center, eye corners, eyeball center, glints) to map coordinates from the user’s eyes plane to coordinates in a viewed plane. Some applications with minimal stimulus require very accurate gaze estimation, such as reading analysis, attention maps, human–computer interaction, among others, and small uncertainties could be very critical to such studies. High-accuracy gaze estimation is essential to describe the actual user’s Point-of-Regard (PoR) truthfully. Some gaze estimation methods can achieve high-accuracy when the gaze error is 0. This work refers to the mapping from gaze estimation onto ground truth as gaze error in pixels or visual angle degrees. On the other hand, precision is the eye-tracking method’s reliability to reproduce the same gaze estimation in successive samples. Accuracy is the average difference between the gaze estimation and the actual stimuli position. Researchers and companies constantly aim to improve eye trackers’ accuracy and precision. Compared to traditional polynomial-based and homography-based gaze estimation methods, the proposed methods increase the number of gaze estimations in the high-accuracy range. We evaluate the improvements achieved with the proposed methods using Gaussian analysis, which defines a range for high-accuracy gaze estimation between − 0. The data analysis uses eye-tracking data from a simulated environment and an experiment with 83 volunteer participants (55 males and 28 females). This paper proposes geometric transformation methods to reshape the eye feature distribution based on the virtual alignment of the eye-camera in the center of the eye’s optical axis. Our experiments show that the eye-camera location combined with the non-coplanarity of the eye plane deforms the eye feature distribution when the eye-camera is far from the eye’s optical axis. Several factors can negatively influence gaze estimation methods when building a commercial or off-the-shelf eye tracker device, including the eye-camera location in uncalibrated setups. This study investigates the influence of the eye-camera location associated with the accuracy and precision of interpolation-based eye-tracking methods. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |