吉林大学学报(信息科学版) ›› 2023, Vol. 41 ›› Issue (1): 180-185.

• • 上一篇    下一篇

基于深度学习的单相机双光源的眼动交互技术

赵培森, 玄玉波, 何 琪   

  1. (吉林大学 通信工程学院 长春 130012)
  • 收稿日期:2022-04-28 出版日期:2023-02-08 发布日期:2023-02-09
  • 通讯作者: 玄玉波(1979— ), 女, 黑龙江牡丹江人, 吉林大学副教授, 主要从事图像处理与并行计算研究, (Tel)86-13596103291(E-mail)xuanyb@ jlu. edu. cn。
  • 作者简介:赵培森(2001— ), 男, 山西阳泉人, 吉林大学本科生, 主要从事信号处理和模式识别研究, ( Tel) 86-13134470508(E-mail)pesionzhao@ 163. com;
  • 基金资助:
    吉林大学大学生创新创业训练计划基金资助项目(202110183174)

Technology of Eye Movement Interaction for Single-Camera and Dual-Light Sources Based on Deep Learning

ZHAO Peisen, XUAN Yubo, HE Qi   

  1. (College of Communication Engineering, Jilin University, Changchun 130012, China)
  • Received:2022-04-28 Online:2023-02-08 Published:2023-02-09

摘要: 为进行高帧率眼动交互, 提出了基于深度学习的单相机双光源识别方法。 该方法运用了反射光斑与视线落点的相关关系, 得到了由眼图到视线落点的映射规律, 构建了人机眼动交互装置并得到高质量数据集, 训练得到了精度与速度较高的视线落点定位模型, 解决了视线估计数学模型复杂、 运算量大的问题。 实验结果表明, 该方法实现了实时识别用户视线落点并进行交互的功能, 可支持心理学实验研究以及虚拟现实应用技术的发展和应用。

关键词: 视线估计, 卷积神经网络, 人机交互

Abstract: In order to realize eye movement interaction with high frame rate, a deep-learning based single-camera dual-light source identification method is proposed. This method uses correlation between the reflected spot and the gaze landing point to obtain the mapping law from the eye image to the gaze. Human-computer eye movement interaction device is constructed, which results in a high-quality dataset, obtains the gaze estimating model with high precision and speed by training, solves the problems of complex mathematical model and large amount of calculation of gaze estimation. It realizes the function of real-time recognition and interaction of users蒺 gaze, supports the development and application of psychological experimental research and virtual reality application technology.

Key words: gaze estimation, convolutional neural network, human-computer interaction

中图分类号: 

  • TP391. 4