Type of Publication

Conference Papers

Date:

2 /

2023

Status

Published

DOI:

10.5220/0011639600003411

Dealing with Overfitting in the Context of Liveness Detection Using FeatherNets with RGB Images

Featured in:

Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods - ICPRAM

Authors:

Miguel Leão and Nuno Gonçalves

Abstract

With the increased use of machine learning for liveness detection solutions comes some shortcomings like overfitting, where the model adapts perfectly to the training set, becoming unusable when used with the testing set, defeating the purpose of machine learning. This paper proposes how to approach overfitting without altering the model used by focusing on the input and output information of the model. The input approach focuses on the information obtained from the different modalities present in the datasets used, as well as how varied the information of these datasets is, not only in number of spoof types but as the ambient conditions when the videos were captured. The output approaches were focused on both the loss function, which has an effect on the actual ”learning”, used on the model which is calculated from the model’s output and is then propagated backwards, and the interpretation of said output to define what predictions are considered as bonafide or spoof. Throughout this work, we were able to reduce the overfitting effect with a difference between the best epoch and the average of the last fifty epochs from 36.57% to 3.63%.

Citation
Miguel Leão and Nuno Gonçalves (2023). Dealing with Overfitting in the Context of Liveness Detection Using FeatherNets with RGB Images. In Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods – ICPRAM; ISBN 978-989-758-626-2; ISSN 2184-4313, SciTePress, pages 62-73. DOI: 10.5220/0011639600003411

Related Content

Researcher Coordinator, VIS TEAM Leader
Researcher
No tagged content to show
No tagged content to show
No tagged content to show

RECENT PUBLICATIONS

MorFacing: A Benchmark for Estimation Face Recognition Robustness to Face Morphing Attacks

Authors: Iurii Medvedev and Nuno Gonçalves
Featured in: IEEE International Joint Conference on Biometrics (IJCB 2024)

Face Liveness Detection Competition (LivDet-Face)

Authors: Lambert Igene, Afzal Hossain, Stephanie Schuckers, Mohammad Zahir Uddin Chowdhury, Humaira Rezaie, Ayden Rollins, Jesse Dykes, Rahul Vijaykumar, Sebastien Marcel, Juan Tapia, Carlos Aravena, Daniel Schulz, Nima Karimian and Anafsheh Adami, Diogo Nunes, João Marcos, Nuno Gonçalves, Lovro Sikošek, Borut Batagelj, Nima Schei, David Pabon, Manuela Tiedemann, Vasiliy Pryadchenko, Aleksandr Alenin, Alhasan Alkhaddour, Anton Pimenov, Artem Tregubov, Igor Avdonin, Maxim Lazantsev and Mikhail Pozigun
Featured in: IEEE International Joint Conference on Biometrics Competitions, 2024

Social NSTransformers: Low-Quality Pedestrian Trajectory Prediction

Authors: Zihan Jiang, Yiqun Ma, Bingyu Shi, Xin Lu, Jian Xing, Nuno Gonçalves and Bo Jin
Featured in: IEEE Transactions on Artificial Intelligence

suggested news

Laser engraving of precious metal artifacts (UniqueMark® deterministic...
UniqueMark® and UniQode® Glitter patent published
Paper about protecting facial recognition systems against morphing...

RECENT PROJECTS

FACING2 – Face Image Understanding
VISUAL-ID – Unique Visual Identities in Graphics, Images and Faces
UniqueMark

Institute of Systems and Robotics Department of Electrical and Computers Engineering University of Coimbra