The Helen Dataset for ICCV 2013
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
data
lib
README.md
demoHelen.m

README.md

Introduction

This page contains the Helen dataset used in the experiment 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

  1. Unzip helen.zip to your folder;
  2. Run demoHelen.

Instructions

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

  • ./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.

References

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

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

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

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},
    }

If you have any question, please feel free to contact Feng Zhou (zhfe99@gmail.com).