Abstract
Extracted data from an eye tracker generally contain inaccurate and noisy signals of spatial gaze position. The noise can affect performance of object selection in gaze-based interactive applications that implement smooth pursuit eye movement to control digital contents. Some signal denoising methods—such as Moving Average Filter, Kalman Filter, and Particle Filter—have been used in a smooth-pursuit-based interactive application to denoise the signal. There is another signal denoising method named Naive Segmented Linear Regression (NSLR) that is able to denoise gaze position signal by segmenting the signal and performing linear regression. Unfortunately, the use of NSLR in a smooth-pursuit-based interactive application is yet to be considered. To fill this research gap, we apply the NSLR signal denoising method and compare its performance on gaze-based object selection against the above-mentioned filters. Experimental results show that gaze-based object selection with Euclidean Distance and Moving Average Filter obtains the best accuracy of 92.37±5.63%, whereas, object selection with Pearson Product Moment Coefficient and Kalman Filter yields the best accuracy of 23.57±2.08%. In future, our results can be used as guidance in the development of smooth-pursuit-based interactive applications.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
