Investigation of the Performance of Different Loss Function Types Within Deep Neural Anchor-Free Object Detectors

Ala’a Alshubbak, Ala’a Alshubbak, Daniel Görges

2024

Abstract

In this paper, an investigation of different IoU loss functions and a spatial attention mechanism within anchor-free object detectors is presented. Two anchor-free dense predictor models are studied: FASF and FCOS models. The models are tested on two different datasets: the benchmark COCO dataset and a small dataset called OPEDD. The results show that some loss functions and using the attention mechanism outperform their original counterparts for both the huge multi-class COCO dataset and the small unity-class dataset of OPEDD. The proposed structure is tested over different backbones: ResNet-50, ResNet-101, and ResNeXt-101. The accuracy of basic models trained over the coco dataset improves by 1.3% and 1.6% mAP for the FSAF and FCOS models based on ResNet-50, respectively. On the other hand, it increases by 2.3% and 15.8% for the same models when trained on the OPEDD dataset. The effect is interpreted using a saliency map.

Paper Citation

in Harvard Style

Alshubbak A. and Görges D. (2024). Investigation of the Performance of Different Loss Function Types Within Deep Neural Anchor-Free Object Detectors. In Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART; ISBN 978-989-758-680-4, SciTePress, pages 401-411. DOI: 10.5220/0012354900003636

in Bibtex Style

@conference{icaart24,
author={Ala’a Alshubbak and Daniel Görges},
title={Investigation of the Performance of Different Loss Function Types Within Deep Neural Anchor-Free Object Detectors},
booktitle={Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART},
year={2024},
pages={401-411},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012354900003636},
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 - Investigation of the Performance of Different Loss Function Types Within Deep Neural Anchor-Free Object Detectors
SN - 978-989-758-680-4
AU - Alshubbak A.
AU - Görges D.
PY - 2024
SP - 401
EP - 411
DO - 10.5220/0012354900003636
PB - SciTePress