Name
Title | ||
|---|---|---|
An eye tracker based on webcam and its preliminary application evaluation in Chinese reading tests |
Abstract | ||
|---|---|---|
There is increasing evidence that eye trackers can be used in neuroscience research to assess cognitive function although expensive prices or low availability still prevent its ubiquitous application. Meanwhile, in tasks such as reaction speed testing in cognitive function assessment, the requirements for spatial performance are less critical than temporal resolution. Therefore, in this report we propose an eye-tracking system based on a common web -camera, to evaluate if it's possible to achieve a decent temporal resolution for eye-tracking tasks in cognitive experiments and then further evaluate it in Chinese reading tests. The performance of the proposed setup are evaluated in comparison experiments including 62 healthy subjects, with a commercialized eye-tracker (SMI RED250) simultaneously deployed. It's found in the results that the webcam-based system showed no significant difference in the recognition accuracy of occurrence and switching of the eye movement events compared with the SMI system though the spatial eye-tracking performance is significantly worse. Then in an application experiment, cognitive-related Chinese reading tests are adapted and EMs of 27 young (18-30 years) and 28 elderly (over 60 years) readers are recorded to study their difference in Chinese reading. In the results there are significant differences in reading-time-per-sentence between young and elderly subjects. Meanwhile, a study on the reading-time-per-sentence difference using reading materials of different difficulties shows that elderly subjects are more sensitive to difficulty changes. These results preliminary demonstrated the feasibility of a Chinese reading test with a web-camera setup as an alternative option when the expensive eye-tracker is not available. |
Year | DOI | Venue |
|---|---|---|
2022 | 10.1016/j.bspc.2022.103521 | BIOMEDICAL SIGNAL PROCESSING AND CONTROL |
Keywords | DocType | Volume |
Eye-tracking, Eye movements, Camera, Deep learning | Journal | 74 |
ISSN | Citations | PageRank |
1746-8094 | 0 | 0.34 |
References | Authors | |
0 | 8 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zhongjun Lin | 1 | 0 | 0.68 |
| Yinghua Liu | 2 | 0 | 0.68 |
| H. Wang | 3 | 67 | 25.36 |
| Zhengji Liu | 4 | 0 | 1.01 |
| Sijin Cai | 5 | 0 | 0.68 |
| Zhicheng Zheng | 6 | 0 | 0.68 |
| Yongjin Zhou | 7 | 17 | 9.89 |
| Xinyu Zhang | 8 | 2 | 1.04 |