A Comprehensive Crossroad Camera Dataset to Improve Traffic Safety of Mobility Aid Users
Ludwig Mohr (Institute of Computer Graphics and Vision, Graz University of Technology),* Nadezda Kirillova (Graz University of Technology), Horst Possegger (Graz University of Technology), Horst Bischof (Graz University of Technology)
The 34th British Machine Vision Conference
Abstract
Improving the safety of traffic participants and reducing the severity of injuries as well as the number of fatalities in the event of accidents is becoming ever more important in the development of vehicles and transportation infrastructure. The most vulnerable group of road users is unquestionably pedestrians, of which people with mobility impairments are especially at risk due to reduced reaction speed or reduced visibility due to smaller silhouettes or unusual postures. Successful strategies for increasing safety by reducing the likelihood of accidents include architectural improvements in planning of pedestrian crossings, as well as advancements in their operation. These strategies can benefit from camera based pedestrian detection systems, yet pedestrians using mobility aids are highly underrepresented in common datasets for object detection and classification, if present at all. To fill this gap and enable researchers to develop methods considering pedestrians in their mobility, we present a novel dataset of pedestrians using mobility aids, together with evaluations of state-of-the-art methods for classification and detection.Video
Citation
@inproceedings{Mohr_2023_BMVC,
author = {Ludwig Mohr and Nadezda Kirillova and Horst Possegger and Horst Bischof},
title = {A Comprehensive Crossroad Camera Dataset to Improve Traffic Safety of Mobility Aid Users},
booktitle = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen, UK, November 20-24, 2023},
publisher = {BMVA},
year = {2023},
url = {https://papers.bmvc2023.org/0743.pdf}
}
Copyright © 2023
The British Machine Vision Conference is organised by