Face Recognition using 3D CNNs

TL;DR

This paper explores using 3D CNNs on video data to improve face recognition robustness under real-world conditions, achieving high accuracy with a new dataset and demonstrating the effectiveness of spatial-temporal feature extraction.

Contribution

It introduces a novel approach of applying 3D CNN architectures to video-based face recognition and develops a new dataset for experimental validation.

Findings

3D CNNs outperform 2D CNNs in real-world face recognition scenarios.

DenseNets achieve 97% accuracy on the CVBL dataset.

Video-based 3D CNN approach shows promising results for surveillance applications.

Abstract

The area of face recognition is one of the most widely researched areas in the domain of computer vision and biometric. This is because, the non-intrusive nature of face biometric makes it comparatively more suitable for application in area of surveillance at public places such as airports. The application of primitive methods in face recognition could not give very satisfactory performance. However, with the advent of machine and deep learning methods and their application in face recognition, several major breakthroughs were obtained. The use of 2D Convolution Neural networks(2D CNN) in face recognition crossed the human face recognition accuracy and reached to 99%. Still, robust face recognition in the presence of real world conditions such as variation in resolution, illumination and pose is a major challenge for researchers in face recognition. In this work, we used video as input to the 3D CNN architectures for capturing both spatial and time domain information from the video for face recognition in real world environment. For the purpose of experimentation, we have developed our own video dataset called CVBL video dataset. The use of 3D CNN for face recognition in videos shows promising results with DenseNets performing the best with an accuracy of 97% on CVBL dataset.