Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
wangguanzhong 817f7cdba3 | 2 years ago | |
---|---|---|
.. | ||
cpp | 2 years ago | |
python | 2 years ago | |
README.md | 2 years ago |
PP-Tracking是基于飞桨深度学习框架的业界首个开源实时跟踪系统。针对实际业务的难点痛点,PP-Tracking内置行人车辆跟踪、跨镜头跟踪、多类别跟踪、小目标跟踪及流量计数等能力与产业应用,同时提供可视化开发界面。模型集成多目标跟踪,目标检测,ReID轻量级算法,进一步提升PP-Tracking在服务器端部署性能。同时支持python,C++部署,适配Linux,Nvidia Jetson多平台环境。
PP-Tracking提供了简洁的可视化界面,无需开发即可实现多种跟踪功能,可以参考PP-Tracking可视化界面使用文档快速上手体验
PP-Tracking集成了多目标跟踪,目标检测,ReID轻量级算法,提升跟踪系统实时性能。多目标跟踪算法基于FairMOT进行优化,实现了服务器端轻量级模型,同时基于不同应用场景提供了针对性的预训练模型。
模型训练评估方法请参考多目标跟踪快速开始
PP-Tracking中提供的多场景预训练模型及导出模型列表如下:
场景 | 数据集 | 精度(MOTA) | NX模型预测速度(FPS) | 配置文件 | 模型权重 | 预测部署模型 |
---|---|---|---|---|---|---|
行人跟踪 | MOT17 | 65.3 | - | 配置文件 | 下载链接 | 下载链接 |
行人小目标跟踪 | VisDrone-pedestrian | 40.5 | - | 配置文件 | 下载链接 | 下载链接 |
车辆跟踪 | BDD100k-vehicle | 32.6 | - | 配置文件 | 下载链接 | 下载链接 |
车辆小目标跟踪 | VisDrone-vehicle | 39.8 | - | 配置文件 | 下载链接 | 下载链接 |
多类别跟踪 | BDD100k | - | - | 配置文件 | 下载链接 | 下载链接 |
多类别小目标跟踪 | VisDrone | 20.4 | - | 配置文件 | 下载链接 | 下载链接 |
更多跟踪模型请参考多目标跟踪模型库
检测模型使用轻量级特色模型PP-PicoDet,具体请参考PP-PicoDet文档
ReID模型使用超轻量骨干网络模型PP-LCNet, 具体请参考PP-LCNet模型介绍
PP-Tracking 使用python预测部署教程请参考PP-Tracking python部署文档
PP-Tracking 使用c++预测部署教程请参考PP-Tracking c++部署文档
PaddleDetection
Python Jupyter Notebook Markdown C++ Java other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》