Deformable Models for Eye Tracking

Martin Vester-Christensen; Denis Leimberg; Bjarne Kjær Ersbøll; Lars Kai Hansen

Deformable Models for Eye Tracking

Martin Vester-Christensen, Denis Leimberg, Bjarne Kjær Ersbøll, Lars Kai Hansen

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research

409 Downloads (Pure)

Abstract

A deformable template method for eye tracking on full face images is presented. The strengths of the method are that it is fast and retains accuracy independently of the resolution. We compare the me\$\backslash\$-thod with a state of the art active contour approach, showing that the heuristic method is more accurate.

Original language	English
Title of host publication	Den 14. Danske Konference i Mønstergenkendelse og Billedanalyse
Publication date	2005
Publication status	Published - 2005

Access to Document

Imm3900

http://www2.imm.dtu.dk/pubdb/p.php?3900

OpenUrl availability

Full text

Cite this

@inproceedings{a36d83bbc89f4a4b80b539272b7a20a8,

title = "Deformable Models for Eye Tracking",

abstract = "A deformable template method for eye tracking on full face images is presented. The strengths of the method are that it is fast and retains accuracy independently of the resolution. We compare the me\$\backslash\$-thod with a state of the art active contour approach, showing that the heuristic method is more accurate.",

keywords = "gaze detection, particle filtering, active contours, deformable templates, eye tracking",

author = "Martin Vester-Christensen and Denis Leimberg and Ersb{\o}ll, {Bjarne Kj{\ae}r} and Hansen, {Lars Kai}",

year = "2005",

language = "English",

booktitle = "Den 14. Danske Konference i M{\o}nstergenkendelse og Billedanalyse",

}

TY - GEN

T1 - Deformable Models for Eye Tracking

AU - Vester-Christensen, Martin

AU - Leimberg, Denis

AU - Ersbøll, Bjarne Kjær

AU - Hansen, Lars Kai

PY - 2005

Y1 - 2005

N2 - A deformable template method for eye tracking on full face images is presented. The strengths of the method are that it is fast and retains accuracy independently of the resolution. We compare the me\$\backslash\$-thod with a state of the art active contour approach, showing that the heuristic method is more accurate.

AB - A deformable template method for eye tracking on full face images is presented. The strengths of the method are that it is fast and retains accuracy independently of the resolution. We compare the me\$\backslash\$-thod with a state of the art active contour approach, showing that the heuristic method is more accurate.

KW - gaze detection

KW - particle filtering

KW - active contours

KW - deformable templates

KW - eye tracking

M3 - Article in proceedings

BT - Den 14. Danske Konference i Mønstergenkendelse og Billedanalyse

ER -