This repository contains scripts to clean the age metadata of the MORPH dataset as mentioned in the paper "Does Face Recognition Accuracy Get Better With Age? Deep Face Matchers Say No".
It also contains the list of subjects' images divided by age range, gender and ethnicity used in the paper.
To get the MORPH dataset click here.
Please cite the papers below if you use the dataset and cleaning steps provided here.
@InProceedings{albiero_wacv_2020,
author={Vítor Albiero and Kevin W. Bowyer and Kushal Vangara and Michael C. King},
title = {Does Face Recognition Accuracy Get Better With Age? Deep Face Matchers Say No},
booktitle={2020 IEEE Winter Conference on Applications of Computer Vision (WACV)},
month = {March},
year = {2020}
}
@INPROCEEDINGS{morph,
author = {Ricanek, Karl and Tesafaye, Tamirat},
booktitle = {International Conference on Automatic Face and Gesture Recognition},
title = {{MORPH: A longitudinal image database of normal adult age-progression}},
year = {2006}
}