On the effect of age perception biases for real age regression

TL;DR

This paper investigates how biases related to gender, ethnicity, makeup, and expression affect deep learning models for age estimation from facial images, proposing an integrated approach that improves both apparent and real age predictions.

Contribution

The main contribution is an end-to-end model that incorporates face attributes for apparent age prediction with an additional loss for real age regression, outperforming existing architectures.

Findings

The proposed model improves age estimation accuracy on the APPA-REAL dataset.

Incorporating face attributes benefits both apparent and real age predictions.

Preliminary application shows potential for gender-based apparent age regression.

Abstract

Automatic age estimation from facial images represents an important task in computer vision. This paper analyses the effect of gender, age, ethnic, makeup and expression attributes of faces as sources of bias to improve deep apparent age prediction. Following recent works where it is shown that apparent age labels benefit real age estimation, rather than direct real to real age regression, our main contribution is the integration, in an end-to-end architecture, of face attributes for apparent age prediction with an additional loss for real age regression. Experimental results on the APPA-REAL dataset indicate the proposed network successfully take advantage of the adopted attributes to improve both apparent and real age estimation. Our model outperformed a state-of-the-art architecture proposed to separately address apparent and real age regression. Finally, we present preliminary results and discussion of a proof of concept application using the proposed model to regress the apparent age of an individual based on the gender of an external observer.