Morph Age

admin



During the battle with Holocaust, Rogue was hurt and Morph, as a means to get her interest, changed to Rogue's son, Charles. Doing the trick, Rogue, Morph and the rest of the team managed to defeat Holocaust and rescue Sabretooth. Shortly afterwards, Morph joined the rest of the X-Men in the final battle with Apocalypse. Characters' age Throughout the publication of the series, there was some dispute about the exact ages of the Animorphs at the time they obtained the ability to morph. However, with the help of various hints in the course of the series, many fans guessed their ages to be approximately thirteen to fourteen (with thirteen being the more likely) at.

Morph Age is a Mac application that lets you create a morphing animation from 2 (or more) faces or objects photos.

Authors: Zhenxing Niu, Mo Zhou, Xinbo Gao, Gang Hua

Download AFAD-Lite: https://github.com/afad-dataset/tarball-lite

Download AFAD-Full: https://github.com/afad-dataset/tarball

Brief Introduction

The Asian Face Age Dataset (AFAD) is a new dataset proposed for evaluating the performance of age estimation, which contains more than 160K facial images and the corresponding age and gender labels. This dataset is oriented to age estimation on Asian faces, so all the facial images are for Asian faces. It is noted that the AFAD is the biggest dataset for age estimation to date. It is well suited to evaluate how deep learning methods can be adopted for age estimation.

Motivation

For age estimation, there are several public datasets for evaluating the performance of a specific algorithm, such as FG-NET [1] (1002 face images), MORPH I (1690 face images), and MORPH II[2] (55,608 face images). Among them, the MORPH II is the biggest public dataset to date. On the other hand, as we know it is necessary to collect a large scale dataset to train a deep Convolutional Neural Network. Therefore, the MORPH II dataset is extensively used to evaluate how deep learning methods can be adopted for age estimation [3][4].

Age

However, the ethnic is very unbalanced for the MORPH II dataset, i.e., it has only less than 1% Asian faces. In order to evaluate the previous methods for age estimation on Asian Faces, the Asian Face Age Dataset (AFAD) was proposed.

Statistics and some samples

There are 164,432 well-labeled photos in the AFAD dataset. It consist of 63,680 photos for female as well as 100,752 photos for male, and the ages range from 15 to 40. The distribution of photo counts for distinct ages are illustrated in the figure above. Some samples are shown in the Figure on the top. Its download link is provided in the 'Download' section.

In addition, we also provide a subset of the AFAD dataset, called AFAD-Lite, which only contains PLACEHOLDER well-labeled photos. It consist of PLACEHOLDER photos for female as well as PLACEHOLDER photos for male, and the ages range from 15 to 40. The distribution of photo counts for distinct ages are illustrated in Fig. PLACEHOLDER. Its download link is also provided in the 'Download' section.

Detailed information for the AFAD dataset

The AFAD dataset is built by collecting selfie photos on a particular social network -- RenRen Social Network (RSN) [5]. The RSN is widely used by Asian students including middle school, high school, undergraduate, and graduate students. Even after leaving from school, some people still access their RSN account to connect with their old classmates. So, the age of the RSN user crosses a wide range from 15-years to more than 40-years old.

Publications

Zhenxing Niu, Mo Zhou, Xinbo Gao, Gang Hua. Ordinal Regression with a Multiple Output CNN for Age Estimation. CVPR, 2016

Download

Please notice that this dataset is made available for academic research purpose only.

1 Dataset: AFAD-LITE

AFAD-LITE is a subset of the complete AFAD dataset, which contains images of 22continuous ages, in the amount of 60K.

2 Dataset: AFAD

AFAD is the complete version of AFAD dataset.

Experiment Results

MethodMAE on MORPH IIMAE on AFAD
BIFS+LSVR [6]4.314.13
BIFS+CCA [7]4.734.40
CNN+LSVR [8]5.135.56
BIFS+OR-SVM [9]4.214.36
BIFS+OHRank [10]3.823.84
OURS []3.273.34

Code

To Be Updated.

FAQ

Morph Agency

To Be Updated.

Reference

  • [1] The fg-net aging database. http://sting.cycollege.ac.cy/alanitis/fgnetaging.html.
  • [2] K. Ricanek and T. Tesafaye. Morph: A longitudinal image database of normal adult age-progression. IEEE International Conference on Automatic Face and Gesture Recognition, pages 341–345, 2015.
  • [3] D. Yi, Z. Lei, and S. Li. Age estimation by multi-scale convolutional network. ACCV, pages 144–158, 2014.
  • [4] X. Wang, R. Guo, and C. Kambhamettu. Deeply-learned feature for age estimation. WACV, pages 534–541, 2015.
  • [5] Renren social network. http://www.renren.com/.
  • [6] G. Guo, G. Mu, Y. Fu, and T. Huang. Human age estimation using bio-inspired features. CVPR, pages 112–119, 2009.
  • [7] G. Guo and G. Mu. Joint estimation of age, gender and ethnicity: Cca vs. pls. FG, pages 1–6, 2013.
  • [8] X. Wang, R. Guo, and C. Kambhamettu. Deeply-learned feature for age estimation. WACV, pages 534–541, 2015.
  • [9] K. Chang, C. Chen, and Y. Hung. A ranking approach for human age estimation based on face images. ICPR, 2010.
  • [10] K. Chang, C. Chen, and Y. Hung. Ordinal hyperplanes ranker with cost sensitivities for age estimation. CVPR, pages 585–592, 2011.
We sometimes want to see how our face changed through ages from a baby until now. More interestingly, we wonder how our face would progress as we get old. The two main age progression categories are child into adult and adult into old age. Combining the two categories is also possible, as a child may become an adult, and then continue to old age. In motion pictures, there are movies in which children physically become adults such as Big, 13 Going on 30. The opposite process of age progress is age regression, which shows how a person gets younger. A series of age progression images are shown in the below figure.

Female Age Progression Videos

Overview of the (fictional) age progression of Bruce Lee, a famous Asian actor.

Applications

Age progression can be used extensively in several applications. For example, it is widely used as a forensics tool by the law enforcement. Age progression can be used to show the likely current appearance of a missing person from a photograph many years old. Our particular interest is to predict what a person would look like when they are younger and older. We also interest in finding what would Bruce Lee would look like if he lives on. Since there was very few Bruce Lee's baby photo, it would be fun to approximate what he would look like when he was young. The entire age progression of Bruce Lee can be achieve using this method with tuned blending factors.
The project is implemented in MATLAB. The software, called Age Progression Manipulator, is developed as the result of the method. Age Progression Manipulator offers a couple of features in which users can experience. Below is a few key features that is available:

Key features

Interactive inputs

You will get to decide how should Bruce looks when he was young or when he would have got older. It's all up to your creativity to morphing the images.

Two blending modes

There's two blending modes you can choose from: shape or color. The shape blending control the facial structure of the morphing images. The color blending controls the skin and hair texture.

Real-time simulation

The program would render your input in realtime. I manage to do so via a fast image morphoing algorithm.

Exportable results

Age

You got to show your result image by printing it. Now go and get on with it!

Downloads

The program is free for non-commericial usage. This download include a single executible file to run in Windows/Linux OS. After download, double click on ageprogression.exe inside the folder to run. If you do not have MATLAB installed on your computer, you first need to download and install the MATLAB Compiler Runtime (MCR) in order to run this program.

PDF file explaining the method

Age Progression ZIP file

Photo Morphing Software

MCRInstaller v.7.8 (for non-MATLAB user)

Compare Morph Age And Fantamorph

  1. Wikipedia. ' Age Progression. 'Article on September 2007. Retrieved December 04, 2009
  2. Arivazhagan, S.; Mumtaj, J.; Ganesan, L. 'Non holonomic' 'Face Recognition Using Multi-Resolution Transform,' Conference on Computational Intelligence and Multimedia Applications, 2007. International Conference on , vol.2, no., pp.301-305, 13-15 Dec. 2007}
  3. Lee D. T., Schachter, B. J,. 'Two algorithms for constructing a Delaunay triangulation,' International Journal of Parallel Programming, Journal on, vol.9, issue 3, pp.219-242, 01 Jun 1980}
  4. Dyn, N. ; Levin, D.; and Rippa, S.Data Dependent Triangulations for Piecewise Linear Interpolation IMA J Numer Anal 10: 137-154.}
  5. Trajkovic, M.; Hedley M.; Fast corner detection, Image and Vision Computing, Volume 16, Issue 2, 20 February 1998, Pages 75-87, ISSN 0262-8856, DOI: 10.1016/S0262-8856(97)00056-5