TD4PWMR
收藏资源简介:
TD4PWMR数据集是由德国明斯特大学数学与计算机科学学院的研究团队创建的,旨在捕捉多样化的行人场景,特别关注在不同环境条件下使用移动辅助设备或承受移动负担的个体。该数据集包含11196张手动标注的热成像图像,涵盖了不同的光照、天气和拥挤的城市环境。该数据集的创建旨在克服热成像的固有挑战,如缺乏色彩和细纹理细节,以及热图像的分辨率通常较低。
TD4PWMR Dataset was created by a research team from the Faculty of Mathematics and Computer Science at the University of Münster, Germany. This dataset is developed to capture diverse pedestrian scenarios, with a particular focus on individuals using mobility assistive devices or bearing mobility-related burdens under various environmental conditions. It contains 11,196 manually annotated thermal imaging images, covering different lighting conditions, weather scenarios, and crowded urban environments. The creation of this dataset aims to address the inherent challenges of thermal imaging, such as the lack of color and fine texture details, as well as the typically lower resolution of thermal images.
YOLO-THERMAL 数据集概述
数据集基本信息
- 名称: YOLO-THERMAL
- 类型: 热成像数据集
- 适用场景: 针对行动受限人群的热成像检测
数据集内容
- 数据来源: TD4PWMR (Thermal Dataset for People with Mobility Restrictions)
- 下载链接:
- Google Drive
- BaiduYun Drive (提取码: 3650)
模型权重
- 预训练模型: yolo-thermal.pt
- 下载链接:
- Google Drive
- BaiduYun Drive (提取码: 3650)
环境配置
- 操作系统: Rocky Linux
- Python版本: 3.10
- PyTorch版本: 2.1.1
- GPU: NVIDIA 2080
安装步骤
-
创建conda环境: bash conda create -n yolo_thermal python=3.10.13 conda activate yolo_thermal
-
安装PyTorch及相关包: bash pip install torch==2.1.1 torchvision==0.16.1 torchaudio==2.1.1 --index-url https://download.pytorch.org/whl/cu118 pip install -r requirements.txt
训练与测试
-
训练脚本: bash python yolo-thermal_train.py
-
评估脚本: bash python yolo-thermal_evaluate.py
引用信息
bibtex @misc{ni2025thermaldetectionpeoplemobility, title={Thermal Detection of People with Mobility Restrictions for Barrier Reduction at Traffic Lights Controlled Intersections}, author={Xiao Ni and Carsten Kuehnel and Xiaoyi Jiang}, year={2025}, eprint={2505.08568}, archivePrefix={arXiv}, primaryClass={cs.CV}, url={https://arxiv.org/abs/2505.08568}, }




