Using Agents and Unsupervised Learning for Counting Objects in Images with Spatial Organization

Eliott Jacopin, Naomie Berda, Léa Courteille, William Grison, Lucas Mathieu, Antoine Cornuéjols, Christine Martin

Abstract

This paper addresses the problem of counting objects from aerial images. Classical approaches either consider the task as a regression problem or view it as a recognition problem of the objects in a sliding window over the images, with, in each case, the need of a lot of labeled images and careful adjustments of the parameters of the learning algorithm. Instead of using a supervised learning approach, the proposed method uses unsupervised learning and an agent-based technique which relies on prior detection of the relationships among objects. The method is demonstrated on the problem of counting plants where it achieves state of the art performance when the objects are well separated and tops the best known performances when the objects overlap. The description of the method underlines its generic nature as it could also be used to count objects organized in a geometric pattern, such as spectators in a performance hall.

Download


Paper Citation


in Harvard Style

Jacopin E., Berda N., Courteille L., Grison W., Mathieu L., Cornuéjols A. and Martin C. (2021). Using Agents and Unsupervised Learning for Counting Objects in Images with Spatial Organization.In Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-484-8, pages 688-697. DOI: 10.5220/0010228706880697


in Bibtex Style

@conference{icaart21,
author={Eliott Jacopin and Naomie Berda and Léa Courteille and William Grison and Lucas Mathieu and Antoine Cornuéjols and Christine Martin},
title={Using Agents and Unsupervised Learning for Counting Objects in Images with Spatial Organization},
booktitle={Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2021},
pages={688-697},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010228706880697},
isbn={978-989-758-484-8},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Using Agents and Unsupervised Learning for Counting Objects in Images with Spatial Organization
SN - 978-989-758-484-8
AU - Jacopin E.
AU - Berda N.
AU - Courteille L.
AU - Grison W.
AU - Mathieu L.
AU - Cornuéjols A.
AU - Martin C.
PY - 2021
SP - 688
EP - 697
DO - 10.5220/0010228706880697