The KITTI Vision Benchmark Suite

Use the detections from PermanceTrack. OC-SORT uses Kalman
Filter for tracking without using appearance. It makes improvement
over the standard SORT by recognizing its limitation in occlusion and
non-linear motion. The inference speed of tracking given off-shelf
detection is 700fps on a i9@3GHz CPU. The method keeps online,
simple and realtime.

detection confidence threshold = 0.6,
IoU matching threshold = 0.3
min_hits = 3

@misc{cao2022observationcentric,
title={Observation-Centric SORT: Rethinking SORT for Robust
Multi-Object Tracking},
author={Jinkun Cao and Xinshuo Weng and Rawal Khirodkar
and Jiangmiao Pang and Kris Kitani},
url={https://arxiv.org/abs/2203.14360},
year={2022},
eprint={2203.14360},
archivePrefix={arXiv},
primaryClass={cs.CV}
}