CNN Based Features Extraction for Age Estimation and Gender Classification

Abstract

This paper proposes automatic age and gender predictions based on feature extraction from human facial images. In contrast to the other traditional methods on the unfiltered benchmarks show their failure to manage large degrees of variation in these types of facial images. In this work, we show that by learning representations through the use of deep convolutional neural networks (CNN), a significant increase in performance can be obtained on these tasks. The novel CNN approach used in this research is made to classify and achieve robust age group and gender classification of unconstrained images. This study has been evaluated and tested on both Essex face dataset and Adience benchmark for gender prediction and age estimation. The results obtained show that the proposed method provide a significant improvement in performance, our model obtains the state-of-the-art performance in both age and gender classification.

Topics & Concepts