An Evaluation of Pre-Trained Models for Feature Extraction in Image Classification

Erick Puls, Matheus Todescato, Joel Carbonera

2024

Abstract

In recent years, we have witnessed a considerable increase in performance in image classification tasks. This performance improvement is mainly due to the adoption of deep learning techniques. Generally, deep learning techniques demand a large set of annotated data, making it challenging when applied to small datasets. Transfer learning strategies have become a promising alternative to overcome these issues in this scenario. This work compares the performance of different pre-trained neural networks for feature extraction in image classification tasks. We evaluated 16 different pre-trained models in four image datasets. Our results demonstrate that the best general performance along the datasets was achieved by CLIP-ViT-B and ViT-H-14, where the CLIP-ResNet50 model had similar performance but with less variability. Therefore, our study provides evidence supporting the choice of models for feature extraction in image classification tasks.

Paper Citation

in Harvard Style

Puls E., Todescato M. and Carbonera J. (2024). An Evaluation of Pre-Trained Models for Feature Extraction in Image Classification. In Proceedings of the 26th International Conference on Enterprise Information Systems - Volume 1: ICEIS; ISBN 978-989-758-692-7, SciTePress, pages 465-476. DOI: 10.5220/0012622300003690

in Bibtex Style

@conference{iceis24,
author={Erick Puls and Matheus Todescato and Joel Carbonera},
title={An Evaluation of Pre-Trained Models for Feature Extraction in Image Classification},
booktitle={Proceedings of the 26th International Conference on Enterprise Information Systems - Volume 1: ICEIS},
year={2024},
pages={465-476},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012622300003690},
isbn={978-989-758-692-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 26th International Conference on Enterprise Information Systems - Volume 1: ICEIS
TI - An Evaluation of Pre-Trained Models for Feature Extraction in Image Classification
SN - 978-989-758-692-7
AU - Puls E.
AU - Todescato M.
AU - Carbonera J.
PY - 2024
SP - 465
EP - 476
DO - 10.5220/0012622300003690
PB - SciTePress