Clustering Facial Attributes: Narrowing the Path From Soft to Hard Biometrics

Andrea F Abate,Fabio Narducci,Silvio Barra,Paola Barra,Cristiano Molinari,Michele Nappi

doi:10.1109/access.2019.2962010

Abstract

Despite the success obtained in face detection and recognition over the last ten years of research, the analysis of facial attributes still represents a trend topic. Keeping the full face recognition aside, exploring the potentials of soft biometric traits, i.e. singular facial traits like the nose, the mouth, the hair and so on, is yet considered a fruitful field of investigation. Being able to infer the identity of an occluded face, e.g. voluntary occluded by sunglasses or accidentally due to environmental factors, can be useful in a wide range of operative fields where user collaboration cannot be considered as an assumption. This especially happens when dealing with forensic scenarios in which is not unusual to have partial face photos or partial fingerprints. In this paper, an unsupervised clustering approach is described. It consists in a neural network model for face attributes recognition based on transfer learning whose goal is grouping faces according to common facial features. Moreover, we use the features collected in each cluster to provide a compact and comprehensive description of the faces belonging to each cluster and deep learning as a mean for task prediction in partially visible faces.

Highlights

Different application scenarios benefit from facial attributes analysis for the purposes of person verification and identification
RELATED WORKS Biometric recognition based on facial traits has been extensively explored over last years and significant robust solutions have been proposed in the literature
Biometric recognition is affected by several critical factors in unconstrained scenarios, which make it a challenging practice to address in an efficient and effective way

Summary

INTRODUCTION

Different application scenarios benefit from facial attributes analysis for the purposes of person verification and identification. In collaborative scenarios the face variations like illumination and pose can be effectively eliminated or greatly reduced, even the detection of the face becomes challenging in unconstrained environments [7] These adverse conditions affecting facial attribute analysis are called PIE-issue, that are:. The approach here proposed relies on CNNs that are used as a facial feature extractor only (to extract features more robust to hand-made features like HOG, Haar, LBP) Such features are used as input for a clustering method which in turns can be used to ease the successive biometric recogntion tasks based on faces. The witness can describe the look of a person and the method can output a set of potential similar subjects

RELATED WORKS

EXPERIMENTAL RESULTS

CONCLUSION