ACADSTAFF UGM
| Judul/Title | [IEEE Access] A Robust Object Selection Technique for Spontaneous Gaze-Based Interaction |
| Penulis/Author | HERLINA (1); Dr.Eng. Ir. Sunu Wibirama, S.T., M.Eng., IPM. (2) |
| Tanggal/Date | 22 2023 |
| Kata Kunci/Keyword | |
| Abstrak/Abstract | Eye tracking provides an alternative input modality to control interactive content on a public display. Instead of using fixational eye movement, spontaneous gaze-based interaction can be performed by implementing smooth pursuit eye movement to trigger a button on the content or to perform an object selection. However, gaze-based object selection based on smooth pursuit eye movement is still inaccurate and unstable. Previous work has not considered gaze drifting that affects accuracy and stability of object selection. To tackle this research gap, we propose a new method for robust gaze-based object selection using Moving Average Filter and Linear Regression. We compared Moving Average Filter with other two well- known signal processing filters used in eye tracking research: Kalman Filter and Particle Filter. We also compared two similarity measures for object selection: Euclidean Distance and Pearson Product-Moment Coefficient. Based on our experimental results, we found significant effects of the Moving Average Filter on the accuracy (p < 0.001) and stability of gaze-based object selection (p < 0.05). In case there are more than one stimuli moving in the same direction, our results suggest that a better performance of object selection can be achieved by considering the absolute distance between the gaze points and the stimulus. In the future, our finding may be used as a guideline for spontaneous gaze-based interaction in a public display. |
| Rumpun Ilmu | Teknologi Informasi |
| Bahasa Asli/Original Language | English |
| Level | Internasional |
| Status |