Hybrid Transfer Learning Using VGG Networks for Robust Ear Biometric Recognition

DOI

10.2991/978-94-6239-723-1_33How to use a DOI?

Keywords

Ear Biometrics; Transfer Learning; VGG Networks; Hybrid Model; Ensemble Fusion; Robust Recognition; Deep Learning; AMI Dataset; EarVN1.0 Dataset

Abstract

Ear biometrics offers a reliable means of personal identification, as the shape of the ear remains largely stable across a person’s lifetime and is unaffected by facial expressions. However, existing deep neural network methods for ear recognition perform poorly in unconstrained, real-world conditions. To address this, we propose a Hybrid Transfer Learning framework based on VGG Networks, which combines the strengths of multiple pre-trained models through score-level ensemble fusion. Four VGG variants - VGG11, VGG13, VGG16, and VGG19 - are individually fine-tuned and evaluated on two benchmark datasets: the controlled AMI Ear Database and the in-the-wild EarVN1.0 dataset. The top-performing pair on each dataset is selected by validation accuracy and combined to form the hybrid model. The proposed ensemble achieves 96.67% recognition accuracy on AMI and 73.96% on EarVN1.0, outperforming every individual VGG variant on both benchmarks.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Mahamadabrar D. Maneri
AU  - Mrunal S. Bewoor
AU  - Ashvini A. Todkar
PY  - 2026
DA  - 2026/07/14
TI  - Hybrid Transfer Learning Using VGG Networks for Robust Ear Biometric Recognition
BT  - Proceedings of the International Conference on Responsible, Risk-aware, and Regulated AI (RRRAI 2026)
PB  - Atlantis Press
SP  - 366
EP  - 375
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-723-1_33
DO  - 10.2991/978-94-6239-723-1_33
ID  - Maneri2026
ER  -