Aging Database Partition for Aging Analysis Evaluation

We used the Morph-II Database [7] for our aging analysis. The database was partitioned into three sets: Set 1, Set 2, and Set 3 (or called others in some of our papers).

The use of this partition of the MORPH database has been in our previous works on aging analysis from different aspects [1-6]. If you want to compare your algorithms with ours, you may use the same partition of the MORPH data.

The file can be downloaded from here:  Morph_AGR_data_segment1.zip

The file is a compressed .mat file. You can uncompress it and then use Matlab to read the struct mat file, from which you can get the corresponding age, gender, ethnicity and id for each face image in the three sets: S1, S1, and Others (S3).

The file is protected by a password. Please email me (guodong dot guo AT mail dot wvu dot edu) to request the password to uncompress the .zip file.

References:

[1]  G-D. Guo and C. Zhang, A study on cross-population age estimation, In the IEEE Conf. on Computer Vision and Pattern Recognition (CVPR14), Columbus, Ohio, June 24-27, 2014. (acceptance rate 29.88%)

[2]  C. Zhang and G-D. Guo, Exploiting unlabeled ages for aging pattern analysis on a large database, In Big Data Computer Vision, in conjunction with CVPR13, Portland, Oregon, June 24, 2013.(Oral)

[3]  G-D. Guo and G. Mu, Joint estimation of age, gender and ethnicity: CCA vs. PLS, Proc. 10th IEEE Int. Conf. Automatic Face and Gesture Recognition (FG13), April, 2013.  (acceptance rate 35.0%)

[4] G-D. Guo and G. Mu, Simultaneous Dimensionality Reduction and Human Age Estimation via Kernel Partial Least Squares Regression, IEEE Conference on Computer Vision and Pattern Recognition (CVPR'11), 657-664, June, 2011.(acceptance rate 26%)

[5] G-D. Guo, G. Mu, Human Age Estimation: What is the Influence Across Race and Gender? IEEE International Workshop on Analysis and Modeling of Faces and Gestures (AMFG'10), June, 2010.

[6] G-D. Guo, G. Mu, A Study of Large-Scale Ethnicity Estimation with Gender and Age Variations, IEEE International Workshop on Analysis and Modeling of Faces and Gestures (AMFG'10), June, 2010.

[7] K. Ricanek and T. Tesafaye. Morph: A longitudinal image database of normal adult age-progression. In IEEE Conf. on AFGR, pages 341C345, 2006. (The MORPH Database)