[BMVC 2023] A Comprehensive Crossroad Camera Dataset of Mobility Aid Users
L. Mohr, N. Kirillova, H. Possegger, H. Bischof
Paper
[TU Graz] Crossroad Camera Dataset - Mobility Aid Users
L. Mohr, N. Kirillova, H. Possegger, H. Bischof
Dataset
[GitHub] YOLOv5 and YOLOv8 modified for YOLO9000-like (but shallow) hierarchical class training
Repository
- Configuration file:
config/opt.py
- Image patch extraction using yolo bounding box annotations:
data_processing/extract_bboxes.py
- Label visualization of extracted patches:
vis_extracted_patches.py
- Bounding box visualization on dataset frames (yolo format):
vis_yolo_labels.py
- Configuration file:
config/opt.py
python main.py
- Configuration file:
config/opt.py
python run.py --model <path/to/model.pth>
--data <path/to/images>
--out <path/where/to/save/predictions>
--vis # Activate prediction visualization
- Configuration file:
config/opt.py
python evaluate.py --model <model name for title>
--gt <path/to/patches/labels.txt>
--pred <path/to/classifier/predictions.txt>
--out <path/where/to/save/predictions>
If you use this code in your work or project, please reference:
@inproceedings{mohr2023mobility
title={{A Comprehensive Crossroad Camera Dataset of Mobility Aid Users}},
author={{Mohr, Ludwig and Kirillova, Nadezda and Possegger, Horst and Bischof, Horst}},
booktitle={{Proceedings of the 34th British Machine Vision Conference ({BMVC})}},
year={2023}
}