pip install -r requirements.txt
pip install -e .
Run main.py
to test the tracker on all videos in the resources
folder.
python tools/main.py cfg/video_track.py
If you want to run it on a webcam, use:
python tools/main.py cfg/video_track.py --cfg-options data.data_root=webcam
If you want to predict on images, use:
python tools/main.py cfg/imgdet.py
If you want to validate the model trained on COCO in the YOLO dataset, refer to the configuration in cfg/imgdet_val.py
.
Define the class names using class_name=[0:'person',1:'car'....]
and pass class_name
to hook.detval
.
- YOLOv5
- YOLOv8
- YOLOv8pose(person)
- YOLOv8seg
- ByteTrack
- Motpy
- NorFair
- OCSort
The ONNX model in the example is exported directly from the official code and can be obtained from the Google Drive.
-
Why did the bounding box (bbox) decoding fail?
Please check the configuration file and ensure that
pipeline.type
is compatible with the model you intend to use.