ahaqu01 / motLinks
基于Yolov5-Deepsort-Fastreid源码,重构了视频行人MOT和行人ReID特征提取代码、接口
☆12Updated 2 years ago
Alternatives and similar repositories for mot
Users that are interested in mot are comparing it to the libraries listed below
Sorting:
- 支持yolov8 Reid行人重识别,支持训练和检测☆16Updated last year
- 基于yoloV5进行多类别+关键检测,关键点检测主要是计算车辆航向角☆16Updated 3 years ago
- 使用yolov8自动标注,运用度量学习metric learning 的ReID算法,实现跨镜头人脸追踪☆10Updated last year
- This repo would give multi-task keypoint detect code based yolov8. The landmarks or keypoints with different classes and numbers can be …☆12Updated 2 years ago
- Yolov5 keras 漂浮物检测 万能运行 数据集制作☆17Updated 4 years ago
- 通过手动标注数据集,训练老人摔倒的模型☆20Updated 2 years ago
- 基于飞桨pphuman中跨镜头跟踪的改进,主要是实现两路推流实时跟踪☆16Updated 2 years ago
- ☆62Updated 3 years ago
- 使用django+pyecharts+PP-Human开发的动态数据大屏, 有人流数据的采集入库, 打架、摔倒等事件警报,口罩检测等实用功能。边缘端版本使用onnx推理提升效率,服务端版本支持视频流推拉☆31Updated 2 years ago
- 使用TensorRT部署SlowFast模型☆23Updated 3 years ago
- ☆43Updated last month
- ☆14Updated 4 years ago
- A project using YoloV8 to detect License Plates☆12Updated 2 years ago
- yolov5+doublehead + MultiLabel+detection☆24Updated 3 years ago
- 使用OpenCV部署yolov5-pose目标检测+人体姿态估计,包含C++和Python两个版本的程序。支持yolov5s,yolov5m,yolov5l☆91Updated 3 years ago
- 使用onnxruntime部署YOWOv2视频动作检测,包含C++和Python两个版本的程序☆28Updated 2 weeks ago
- Real-Time AI Posture Correction for Powerlifting Exercises Using YOLOv5 and MediaPipe☆16Updated 2 months ago
- ☆85Updated 4 years ago
- ☆18Updated 2 years ago
- 多类别多目标跟踪YoloV5+sort/deepsort/bytetrack/BotSort/motdt☆73Updated 2 years ago
- yolov5 with more backbone☆19Updated 3 years ago
- 该项目可用于做yolov5+reid识别,可用于做跨视频人员检测☆46Updated 8 months ago
- RT-DETRv2 tensorrt C++ 部署☆21Updated last year
- A Python implementation of Yolov5 to detect whether peaple smoking in Jetson Xavier nx and Jetson nano☆63Updated 3 years ago
- ☆19Updated 4 years ago
- 基于rknn的yolov5的cpp实现,包含各种依赖库,是一个完整工程,可直接编译运行☆21Updated 3 years ago
- 使用ONNXRuntime部署YOLOV7人头检测,包含C++和Python两个版本的程序☆31Updated 3 years ago
- ☆15Updated 5 years ago
- yolov5模型训练后量化代码☆19Updated 4 years ago
- 重识别完整pipeline。支持视频/图像/遍历文件夹。reid模型采用2024行人重识别指标最高模型.行人检测采用yolov8,行人跟踪采用botsort。检索系统采用faiss。☆29Updated last year