FireFly: A Synthetic Dataset for Ember Detection in Wildfire
Abstract
This paper presents "FireFly", a synthetic dataset for ember detection created using Unreal Engine 4 (UE4), designed to overcome the current lack of ember-specific training resources. To create the dataset, we present a tool that allows the automated generation of the synthetic labeled dataset with adjustable parameters, enabling data diversity from various environmental conditions, making the dataset both diverse and customizable based on user requirements. We generated a total of 19,273 frames that have been used to evaluate FireFly on four popular object detection models. Further to minimize human intervention, we leveraged a trained model to create a semi-automatic labeling process for real-life ember frames. Moreover, we demonstrated an up to 8.57% improvement in mean Average Precision (mAP) in real-world wildfire scenarios compared to models trained exclusively on a small real dataset.
Cite
Text
Hu et al. "FireFly: A Synthetic Dataset for Ember Detection in Wildfire." IEEE/CVF International Conference on Computer Vision Workshops, 2023. doi:10.1109/ICCVW60793.2023.00406Markdown
[Hu et al. "FireFly: A Synthetic Dataset for Ember Detection in Wildfire." IEEE/CVF International Conference on Computer Vision Workshops, 2023.](https://mlanthology.org/iccvw/2023/hu2023iccvw-firefly/) doi:10.1109/ICCVW60793.2023.00406BibTeX
@inproceedings{hu2023iccvw-firefly,
title = {{FireFly: A Synthetic Dataset for Ember Detection in Wildfire}},
author = {Hu, Yue and Ye, Xinan and Liu, Yifei and Kundu, Souvik and Datta, Gourav and Mutnuri, Srikar and Asavisanu, Namo and Ayanian, Nora and Psounis, Konstantinos and Beerel, Peter A.},
booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
year = {2023},
pages = {3767-3771},
doi = {10.1109/ICCVW60793.2023.00406},
url = {https://mlanthology.org/iccvw/2023/hu2023iccvw-firefly/}
}