标题:Block-Wise Gaze Estimation Based on Binocular Images
作者:Wu, Xuemei; Li, Jing; Wu, Qiang; Sun, Jiande; Yan, Hua
通讯作者:Sun, Jiande;Sun, JD
作者机构:[Wu, Xuemei; Wu, Qiang] Shandong Univ, Sch Informat Sci & Engn, Jinan, Shandong, Peoples R China.; [Li, Jing] Shandong Management Univ, Sch Mech & E 更多
会议名称:8th Pacific-Rim Symposium on Image and Video Technology (PSIVT)
会议日期:NOV 20-24, 2017
来源:IMAGE AND VIDEO TECHNOLOGY (PSIVT 2017)
出版年:2018
卷:10749
页码:477-487
DOI:10.1007/978-3-319-75786-5_38
关键词:Gaze estimation; Gaze block; Appearance-based; Eye image; Convolutional; neural network (CNN)
摘要:Appearance-based gaze estimation methods have been proved to be highly effective. Different from the previous methods that estimate gaze direction based on left or right eye image separately, we propose a binocular-image based gaze estimation method. Considering the challenges in estimating the precise gaze points via regression models, we estimate the block-wise gaze position by classifying the binocular images via convolutional neural network (CNN) in the proposed method. We divide the screen of the desktop computer into 2 x 3 and 6 x 9 blocks respectively, label the binocular images with their corresponding gazed block positions, train a convolutional neural network model to classify the eye images according to their labels, and estimate the gazed block through the CNN-based classification. The experimental results demonstrate that the proposed gaze estimation method based on binocular images can reach higher accuracy than those based on monocular images. And the proposed method shows its great potential in practical touch screen-based applications.
收录类别:CPCI-S;EI;SCOPUS
资源类型:会议论文;期刊论文
原文链接:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85042520818&doi=10.1007%2f978-3-319-75786-5_38&partnerID=40&md5=5625e45d5597f23ea58e0fb1562cc6f9
TOP