Scene Detection in De Boer Historical Photo Collection

Melvin Wevers

Abstract

This paper demonstrates how transfer learning can be used to improve scene detection applied to a historical press photo collection. After applying transfer learning to a pre-trained Places-365 ResNet-50 model, we achieve a Top-1 accuracy of .68 and a Top-5 accuracy of .89 on our data set, which consists of 132 categories. In addition to describing our annotation and training strategy, we also reflect on the use of transfer learning and the evaluation of computer vision models for heritage institutes.

Download


Paper Citation


in Harvard Style

Wevers M. (2021). Scene Detection in De Boer Historical Photo Collection.In Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 1: ARTIDIGH, ISBN 978-989-758-484-8, pages 601-610. DOI: 10.5220/0010288206010610


in Bibtex Style

@conference{artidigh21,
author={Melvin Wevers},
title={Scene Detection in De Boer Historical Photo Collection},
booktitle={Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 1: ARTIDIGH,},
year={2021},
pages={601-610},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010288206010610},
isbn={978-989-758-484-8},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Agents and Artificial Intelligence - Volume 1: ARTIDIGH,
TI - Scene Detection in De Boer Historical Photo Collection
SN - 978-989-758-484-8
AU - Wevers M.
PY - 2021
SP - 601
EP - 610
DO - 10.5220/0010288206010610