Helen Dataset for Facial Landmark Localization

DOWNLOAD

Introduction

This page contains the Helen dataset used in the experiments of exemplar-based graph matching (EGM)^[1] for facial landmark detection. The original Helen dataset^[2] adopts a highly detailed annotation. We re-labeled 348 images with the same 29 landmarks as the LFPW dataset^[3]. In addition, we provide MATLAB interface code for loading and visualizing the facial landmarks on images.

Installation

                    >> cd helen
                    >> ls
                    demoHelen.m README.md data lib
                    >> demoHelen
                  

Package Content

The package of helen.zip contains following folders and files:

data:	This folder contains the images and labels of the Helen dataset.
lib:	This folder contains some library functions for visualizing facial landmark on images.
demoHelen.m:	An interface demo for visualizing image and facial landmarks.

Change Log

2013-10-20:

Add interface code, images and labels.

References

[1]

Exemplar-based Graph Matching for Robust Facial Landmark Localization
IEEE International Conference on Computer Vision (ICCV), 2013
F. Zhou, J. Brandt and Z. Lin
[2]

Interactive Facial Feature Localization
European Conference Computer Vision (ECCV), 2012
V. Le, J. Brandt, L. Bourdev, Z. Lin and T. Huang
[3]

Localizing Parts of Faces Using a Consensus of Exemplars
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011
P. N. Belhumeur, D. W. Jacobs, D. J. Kriegman and N. Kumar

Copyright

This software is free for use in research projects. If you publish results obtained using this software, please use this citation:

@inproceedings{ZhouBL13, author = {F. Zhou, J. Brandt and Z. Lin}, title = {Exemplar-based Graph Matching for Robust Facial Landmark Localization}, booktitle = {IEEE International Conference on Computer Vision (ICCV)}, year = {2013}, }

Contributing back bugfixes and improvements is polite and encouraged. If you have any question, feel free to contact Feng Zhou.