Autor: |
Chien-Hui Liu, Wen-Chung Kao, Sheng-Ju Wu, Shih-Yao Yin, Wei-Te Chang |
Rok vydání: |
2013 |
Předmět: |
|
Zdroj: |
2013 International Conference on System Science and Engineering (ICSSE). |
DOI: |
10.1109/icsse.2013.6614676 |
Popis: |
Some gaze tracking systems have been proposed in the field, but very few of them can be directly applied in normal lighting environments. Many image imperfection issues are difficult to be dealt with so that a typical system is usually equipped with a strong infrared ray (IR) illuminant. The user experience is not so good even if the IR illuminant is invisible. In this paper, we present a high speed (240 frames/s) gaze tracking system operating in a normal office lighting condition. A set of image processing algorithms to deal with the issues of poor image contrast and high random noise have been presented. The experimental result shows that the system can achieve the accuracy of 1° and 1.7° for the horizontal and vertical coordinates of the detected iris center, respectively. |
Databáze: |
OpenAIRE |
Externí odkaz: |
|