基于人工神经网络动态标定算法的低成本视线追踪系统

Xiang Zhou Wang; Xin Zhang; Shu Hua Zheng

doi:10.15918/j.tbit1001-0645.2018.12.009

基于人工神经网络动态标定算法的低成本视线追踪系统

Translated title of the contribution: A Low-Cost Eye-Gaze Tracking System Based on Artificial Neural Network Dynamic Calibration Algorithm

Xiang Zhou Wang, Xin Zhang, Shu Hua Zheng^*

^*Corresponding author for this work

School of Automation

Beijing Institute of Technology

Research output: Contribution to journal › Article › peer-review

4 Citations (Scopus)

Abstract

In order to reduce the cost of eye-gaze tracking system and simplify the complexity of the calibration algorithm, a low-cost eye-gaze tracking system was developed.The Haar-like feature and skin color combination algorithm were used to detect the human face.The active appearance model (AAM) algorithm and the optical flow method were used to locate and track the face feature points.And the pupil center was detected by the gradient vector method.An artificial neural network dynamic calibration algorithm was proposed to improve the tracking accuracy and robustness.Experiments show that the eye-gaze tracking system not only has better robustness, but also has higher precision.The average error of the system is 1.34° at head rest, and 3.26° at head movement.

Translated title of the contribution	A Low-Cost Eye-Gaze Tracking System Based on Artificial Neural Network Dynamic Calibration Algorithm
Original language	Chinese (Traditional)
Pages (from-to)	1263-1268
Number of pages	6
Journal	Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology
Volume	38
Issue number	12
DOIs	https://doi.org/10.15918/j.tbit1001-0645.2018.12.009
Publication status	Published - 1 Dec 2018

Access to Document

10.15918/j.tbit1001-0645.2018.12.009

Cite this

@article{9574b684c7844565a659890eaf0a7e03,

title = "基于人工神经网络动态标定算法的低成本视线追踪系统",

abstract = "In order to reduce the cost of eye-gaze tracking system and simplify the complexity of the calibration algorithm, a low-cost eye-gaze tracking system was developed.The Haar-like feature and skin color combination algorithm were used to detect the human face.The active appearance model (AAM) algorithm and the optical flow method were used to locate and track the face feature points.And the pupil center was detected by the gradient vector method.An artificial neural network dynamic calibration algorithm was proposed to improve the tracking accuracy and robustness.Experiments show that the eye-gaze tracking system not only has better robustness, but also has higher precision.The average error of the system is 1.34° at head rest, and 3.26° at head movement.",

keywords = "Active appearance model(AAM), Artificial neural network, Dynamic calibration algorithm, Gradient vector",

author = "Wang, {Xiang Zhou} and Xin Zhang and Zheng, {Shu Hua}",

year = "2018",

month = dec,

day = "1",

doi = "10.15918/j.tbit1001-0645.2018.12.009",

language = "繁体中文",

volume = "38",

pages = "1263--1268",

journal = "Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology",

issn = "1001-0645",

publisher = "Beijing Institute of Technology",

number = "12",

}

TY - JOUR

T1 - 基于人工神经网络动态标定算法的低成本视线追踪系统

AU - Wang, Xiang Zhou

AU - Zhang, Xin

AU - Zheng, Shu Hua

PY - 2018/12/1

Y1 - 2018/12/1

N2 - In order to reduce the cost of eye-gaze tracking system and simplify the complexity of the calibration algorithm, a low-cost eye-gaze tracking system was developed.The Haar-like feature and skin color combination algorithm were used to detect the human face.The active appearance model (AAM) algorithm and the optical flow method were used to locate and track the face feature points.And the pupil center was detected by the gradient vector method.An artificial neural network dynamic calibration algorithm was proposed to improve the tracking accuracy and robustness.Experiments show that the eye-gaze tracking system not only has better robustness, but also has higher precision.The average error of the system is 1.34° at head rest, and 3.26° at head movement.

AB - In order to reduce the cost of eye-gaze tracking system and simplify the complexity of the calibration algorithm, a low-cost eye-gaze tracking system was developed.The Haar-like feature and skin color combination algorithm were used to detect the human face.The active appearance model (AAM) algorithm and the optical flow method were used to locate and track the face feature points.And the pupil center was detected by the gradient vector method.An artificial neural network dynamic calibration algorithm was proposed to improve the tracking accuracy and robustness.Experiments show that the eye-gaze tracking system not only has better robustness, but also has higher precision.The average error of the system is 1.34° at head rest, and 3.26° at head movement.

KW - Active appearance model(AAM)

KW - Artificial neural network

KW - Dynamic calibration algorithm

KW - Gradient vector

UR - http://www.scopus.com/inward/record.url?scp=85062991569&partnerID=8YFLogxK

U2 - 10.15918/j.tbit1001-0645.2018.12.009

DO - 10.15918/j.tbit1001-0645.2018.12.009

M3 - 文章

AN - SCOPUS:85062991569

SN - 1001-0645

VL - 38

SP - 1263

EP - 1268

JO - Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

JF - Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

IS - 12

ER -

基于人工神经网络动态标定算法的低成本视线追踪系统

Abstract

Access to Document

Other files and links

Fingerprint

Cite this