Remote person authentication in different scenarios based on gait and face in front view

To predict criminal acts and to assure more security, the biometric remote recognition of people has lately been getting much interest among researchers. We propose in this paper to use biometric modalities that may be acquired remotely, which are the gait and the face. The gait is explored at 11 di...

Full description

Saved in:
Bibliographic Details
Main Authors: El Kissi Ghalleb, Asma, Ben Amara, Najoua Essoukri
Format: Conference Proceeding
Language:English
Subjects:
angle of view
Cameras
data fusion
Face
Face recognition
Feature extraction
gait
Image color analysis
Legged locomotion
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To predict criminal acts and to assure more security, the biometric remote recognition of people has lately been getting much interest among researchers. We propose in this paper to use biometric modalities that may be acquired remotely, which are the gait and the face. The gait is explored at 11 different angles of view with different styles of clothes using the CASIA Gait Datasets A and B. For the front view, we fuse the gait with hard and soft facial biometrics. Tested on both public databases, the gait-based recognition has yielded interesting results in different cases compared to the existing results in the literature. The system based on the fusion of the gait with the face has led to better results.
ISSN:2474-0446
DOI:10.1109/SSD.2017.8167008