Towards accurate and robust cross-ratio based gaze trackers through learning from simulation

Jia Bin Huang, Qin Cai, Zicheng Liu, Narendra Ahuja, Zhengyou Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

remote gaze estimation using a single camera in an uncalibrated setup by exploiting invariance of a plane projectivity. Unfortunately, due to several simplification assumptions, the performance of CR-based eye gaze trackers decays significantly as the subject moves away from the calibration position. In this paper, we introduce an adaptive homography mapping for achieving gaze prediction with higher accuracy at the calibration position and more robustness under head movements. This is achieved with a learningbased method for compensating both spatially-varying gaze errors and head pose dependent errors simultaneously in a unified framework. The model of adaptive homography is trained offline using simulated data, saving a tremendous amount of time in data collection. We validate the effectiveness of the proposed approach using both simulated and real data from a physical setup. We show that our method compares favorably against other state-of-the-art CR based methods.

Original languageEnglish (US)
Title of host publicationProceedings of the Symposium on Eye Tracking Research and Applications, ETRA 2014
PublisherAssociation for Computing Machinery
Pages75-82
Number of pages8
ISBN (Print)9781450327510
DOIs
StatePublished - Jan 1 2014
Event8th Symposium on Eye Tracking Research and Applications, ETRA 2014 - Safety Harbor, FL, United States
Duration: Mar 26 2014Mar 28 2014

Publication series

NameEye Tracking Research and Applications Symposium (ETRA)

Other

Other8th Symposium on Eye Tracking Research and Applications, ETRA 2014
CountryUnited States
CitySafety Harbor, FL
Period3/26/143/28/14

Fingerprint

Learning
Calibration
Invariance
Head Movements
Cameras
Head

Keywords

  • Cross-ratio
  • Eye gaze tracking
  • Simulation

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Human-Computer Interaction
  • Ophthalmology
  • Sensory Systems

Cite this

Huang, J. B., Cai, Q., Liu, Z., Ahuja, N., & Zhang, Z. (2014). Towards accurate and robust cross-ratio based gaze trackers through learning from simulation. In Proceedings of the Symposium on Eye Tracking Research and Applications, ETRA 2014 (pp. 75-82). (Eye Tracking Research and Applications Symposium (ETRA)). Association for Computing Machinery. https://doi.org/10.1145/2578153.2578162

Towards accurate and robust cross-ratio based gaze trackers through learning from simulation. / Huang, Jia Bin; Cai, Qin; Liu, Zicheng; Ahuja, Narendra; Zhang, Zhengyou.

Proceedings of the Symposium on Eye Tracking Research and Applications, ETRA 2014. Association for Computing Machinery, 2014. p. 75-82 (Eye Tracking Research and Applications Symposium (ETRA)).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Huang, JB, Cai, Q, Liu, Z, Ahuja, N & Zhang, Z 2014, Towards accurate and robust cross-ratio based gaze trackers through learning from simulation. in Proceedings of the Symposium on Eye Tracking Research and Applications, ETRA 2014. Eye Tracking Research and Applications Symposium (ETRA), Association for Computing Machinery, pp. 75-82, 8th Symposium on Eye Tracking Research and Applications, ETRA 2014, Safety Harbor, FL, United States, 3/26/14. https://doi.org/10.1145/2578153.2578162
Huang JB, Cai Q, Liu Z, Ahuja N, Zhang Z. Towards accurate and robust cross-ratio based gaze trackers through learning from simulation. In Proceedings of the Symposium on Eye Tracking Research and Applications, ETRA 2014. Association for Computing Machinery. 2014. p. 75-82. (Eye Tracking Research and Applications Symposium (ETRA)). https://doi.org/10.1145/2578153.2578162
Huang, Jia Bin ; Cai, Qin ; Liu, Zicheng ; Ahuja, Narendra ; Zhang, Zhengyou. / Towards accurate and robust cross-ratio based gaze trackers through learning from simulation. Proceedings of the Symposium on Eye Tracking Research and Applications, ETRA 2014. Association for Computing Machinery, 2014. pp. 75-82 (Eye Tracking Research and Applications Symposium (ETRA)).
@inproceedings{a3747cfc49104f9ab6823ed451f9b9dd,
title = "Towards accurate and robust cross-ratio based gaze trackers through learning from simulation",
abstract = "remote gaze estimation using a single camera in an uncalibrated setup by exploiting invariance of a plane projectivity. Unfortunately, due to several simplification assumptions, the performance of CR-based eye gaze trackers decays significantly as the subject moves away from the calibration position. In this paper, we introduce an adaptive homography mapping for achieving gaze prediction with higher accuracy at the calibration position and more robustness under head movements. This is achieved with a learningbased method for compensating both spatially-varying gaze errors and head pose dependent errors simultaneously in a unified framework. The model of adaptive homography is trained offline using simulated data, saving a tremendous amount of time in data collection. We validate the effectiveness of the proposed approach using both simulated and real data from a physical setup. We show that our method compares favorably against other state-of-the-art CR based methods.",
keywords = "Cross-ratio, Eye gaze tracking, Simulation",
author = "Huang, {Jia Bin} and Qin Cai and Zicheng Liu and Narendra Ahuja and Zhengyou Zhang",
year = "2014",
month = "1",
day = "1",
doi = "10.1145/2578153.2578162",
language = "English (US)",
isbn = "9781450327510",
series = "Eye Tracking Research and Applications Symposium (ETRA)",
publisher = "Association for Computing Machinery",
pages = "75--82",
booktitle = "Proceedings of the Symposium on Eye Tracking Research and Applications, ETRA 2014",

}

TY - GEN

T1 - Towards accurate and robust cross-ratio based gaze trackers through learning from simulation

AU - Huang, Jia Bin

AU - Cai, Qin

AU - Liu, Zicheng

AU - Ahuja, Narendra

AU - Zhang, Zhengyou

PY - 2014/1/1

Y1 - 2014/1/1

N2 - remote gaze estimation using a single camera in an uncalibrated setup by exploiting invariance of a plane projectivity. Unfortunately, due to several simplification assumptions, the performance of CR-based eye gaze trackers decays significantly as the subject moves away from the calibration position. In this paper, we introduce an adaptive homography mapping for achieving gaze prediction with higher accuracy at the calibration position and more robustness under head movements. This is achieved with a learningbased method for compensating both spatially-varying gaze errors and head pose dependent errors simultaneously in a unified framework. The model of adaptive homography is trained offline using simulated data, saving a tremendous amount of time in data collection. We validate the effectiveness of the proposed approach using both simulated and real data from a physical setup. We show that our method compares favorably against other state-of-the-art CR based methods.

AB - remote gaze estimation using a single camera in an uncalibrated setup by exploiting invariance of a plane projectivity. Unfortunately, due to several simplification assumptions, the performance of CR-based eye gaze trackers decays significantly as the subject moves away from the calibration position. In this paper, we introduce an adaptive homography mapping for achieving gaze prediction with higher accuracy at the calibration position and more robustness under head movements. This is achieved with a learningbased method for compensating both spatially-varying gaze errors and head pose dependent errors simultaneously in a unified framework. The model of adaptive homography is trained offline using simulated data, saving a tremendous amount of time in data collection. We validate the effectiveness of the proposed approach using both simulated and real data from a physical setup. We show that our method compares favorably against other state-of-the-art CR based methods.

KW - Cross-ratio

KW - Eye gaze tracking

KW - Simulation

UR - http://www.scopus.com/inward/record.url?scp=84899693799&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899693799&partnerID=8YFLogxK

U2 - 10.1145/2578153.2578162

DO - 10.1145/2578153.2578162

M3 - Conference contribution

AN - SCOPUS:84899693799

SN - 9781450327510

T3 - Eye Tracking Research and Applications Symposium (ETRA)

SP - 75

EP - 82

BT - Proceedings of the Symposium on Eye Tracking Research and Applications, ETRA 2014

PB - Association for Computing Machinery

ER -