This repo is a packaged version of the ByteTrack algorithm.
ByteTrack is a multi-object tracking computer vision model. Using ByteTrack, you can allocate IDs for unique objects in a video for use in tracking objects.
To install the library, run the following command:
pip install git+https://github.com/artefactory-fr/bytetrack.git@mainTo install a specific version, run the following command:
pip install git+https://github.com/artefactory-fr/bytetrack.git@x.y.zfrom bytetracker import BYTETracker
tracker = BYTETracker(args)
for frame_id, image_filename in enumerate(frames):
img = cv2.imread(image_filename)
detections = your_model.predict(img)
tracked_objects = tracker.update(detections, frame_id)Copyright (c) 2022 Kadir Nar
ByteTrack is licensed under the MIT License. See the LICENSE file and the ByteTrack repository for more information.
This is a fork of bytetrack-pip.
@article{zhang2022bytetrack,
title={ByteTrack: Multi-Object Tracking by Associating Every Detection Box},
author={Zhang, Yifu and Sun, Peize and Jiang, Yi and Yu, Dongdong and Weng, Fucheng and Yuan, Zehuan and Luo, Ping and Liu, Wenyu and Wang, Xinggang},
booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},
year={2022}
}