Class Anchor Margin Loss for Content-Based Image Retrieval

Alexandru Ghita, Radu Tudor Ionescu

2024

Abstract

Loss functions play a major role in influencing the effectiveness of neural networks in content-based image retrieval (CBIR). Existing loss functions can be categorized into metric learning and statistical learning. Metric learning often lacks efficiency due to pair mining, while statistical learning does not yield compact features. To this end, we introduce a novel repeller-attractor loss based on metric learning, which directly optimizes the L2 metric, without pair generation. Our novel loss comprises three terms: one to ensure features are attracted to class anchors, one that enforces anchor separability, and one that prevents anchor collapse. We evaluate our objective, applied to both convolutional and transformer architectures, on CIFAR-100, Food-101, SVHN, and ImageNet-200, showing that it outperforms existing functions in CBIR.

Paper Citation

in Harvard Style

Ghita A. and Tudor Ionescu R. (2024). Class Anchor Margin Loss for Content-Based Image Retrieval. In Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART; ISBN 978-989-758-680-4, SciTePress, pages 848-853. DOI: 10.5220/0012400500003636

in Bibtex Style

@conference{icaart24,
author={Alexandru Ghita and Radu Tudor Ionescu},
title={Class Anchor Margin Loss for Content-Based Image Retrieval},
booktitle={Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART},
year={2024},
pages={848-853},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012400500003636},
isbn={978-989-758-680-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART
TI - Class Anchor Margin Loss for Content-Based Image Retrieval
SN - 978-989-758-680-4
AU - Ghita A.
AU - Tudor Ionescu R.
PY - 2024
SP - 848
EP - 853
DO - 10.5220/0012400500003636
PB - SciTePress