1chimaruGin / SimpleTracker

- Multi Object Tracker

Simple Object Tracking

Multi-Object Tracking with YOLOv5 and Norfair tracker

Requirements

• torch >= 1.10.1
• torchvision >= 0.11.1

Other requirements can be installed with pip install -r requirements.txt.

Clone the repository recursively:

$ git clone --recursive [e-mail]:1chimaruGin/SimpleTracker.git

Then download a YOLO model's weight from YOLOv5 and place it in weights.

Tracking

You can change the class you want to track. Check the class id here.

$ python tracker.py --source /path/to/video --model weights/yolov5m6.pt --classes 0

References

• https://github.com/ultralytics/yolov5
• https://github.com/sithu31296/simple-object-tracking
• https://github.com/tryolabs/norfair

Citations

@software{joaquin_alori_2022_6471349,
  author       = {Joaquín Alori and
                  Alan Descoins and
                  KotaYuhara and
                  David and
                  facundo-lezama and
                  Braulio Ríos and
                  fatih and
                  shafu.eth and
                  Agustín Castro and
                  David Huh},
  title        = {tryolabs/norfair: v0.4.3},
  month        = apr,
  year         = 2022,
  publisher    = {Zenodo},
  version      = {v0.4.3},
  doi          = {10.5281/zenodo.6471349},
  url          = {https://doi.org/10.5281/zenodo.6471349}
}