noahcao/OC_SORT
A Kalman filter-based multi-object tracker for video that remains robust during occlusion and non-linear motion.
Not currently ranked — collecting fresh signals.
star history
OC-SORT (Observation-Centric SORT) is a pure motion-model-based multi-object tracker designed to improve tracking robustness in crowded scenes and scenarios with non-linear object motion. It recognizes and addresses limitations in the original SORT algorithm and Kalman filter approaches. The system integrates with different detectors and matching modules, including appearance similarity, while remaining simple, online, and real-time.
Frequently asked
- What is noahcao/OC_SORT?
- A Kalman filter-based multi-object tracker for video that remains robust during occlusion and non-linear motion.
- Is OC_SORT open source?
- Yes — noahcao/OC_SORT is open source, released under the MIT license.
- What language is OC_SORT written in?
- noahcao/OC_SORT is primarily written in Python.
- How popular is OC_SORT?
- noahcao/OC_SORT has 1.1k stars on GitHub.
- Where can I find OC_SORT?
- noahcao/OC_SORT is on GitHub at https://github.com/noahcao/OC_SORT.