Nearest Neighbor-Based Data Denoising for Deep Metric Learning

George Galanakis, George Galanakis, Xenophon Zabulis, Antonis Argyros, Antonis Argyros

2024

Abstract

The effectiveness of supervised deep metric learning relies on the availability of a correctly annotated dataset, i.e., a dataset where images are associated with correct class labels. The presence of incorrect labels in a dataset disorients the learning process. In this paper, we propose an approach to combat the presence of such label noise in datasets. Our approach operates online, during training and on the batch level. It examines the neighborhood of samples, considers which of them are noisy and eliminates them from the current training step. The neighborhood is established using features obtained from the entire dataset during previous training epochs and therefore is updated as the model learns better data representations. We evaluate our approach using multiple datasets and loss functions, and demonstrate that it performs better or comparably to the competition. At the same time, in contrast to the competition, it does not require knowledge of the noise contamination rate of the employed datasets.

Paper Citation

in Harvard Style

Galanakis G., Zabulis X. and Argyros A. (2024). Nearest Neighbor-Based Data Denoising for Deep Metric Learning. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 595-603. DOI: 10.5220/0012383000003660

in Bibtex Style

@conference{visapp24,
author={George Galanakis and Xenophon Zabulis and Antonis Argyros},
title={Nearest Neighbor-Based Data Denoising for Deep Metric Learning},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP},
year={2024},
pages={595-603},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012383000003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP
TI - Nearest Neighbor-Based Data Denoising for Deep Metric Learning
SN - 978-989-758-679-8
AU - Galanakis G.
AU - Zabulis X.
AU - Argyros A.
PY - 2024
SP - 595
EP - 603
DO - 10.5220/0012383000003660
PB - SciTePress