Home

Awesome

NV-VOT211 Static Badge

The official implementation for the ACCV 2024 paper [NT-VOT211: A Large-Scale Benchmark for Night-time Visual Object Tracking] introc-1

Yu Liu, Arif Mahmood, Muhammad Haris Khan

Abstract

Many current visual object tracking benchmarks such as OTB100, NfS, UAV123, LaSOT, and GOT-10K, predominantly contain day-time scenarios while the challenges posed by the night-time has been less investigated. It is primarily because of the lack of a large-scale, well-annotated night-time benchmark for rigorously evaluating tracking algorithms. To this end, this paper presents NT-VOT211, a new benchmark tailored for evaluating visual object tracking algorithms in the challenging night-time conditions. NT-VOT211 consists of 211 diverse videos, offering 211,000 well-annotated frames with 8 attributes including camera motion, deformation, fast motion, motion blur, tiny target, distractors, occlusion and out-of-view.

News(last update 2024-10-14)

:fire::fire::fire: <sub>We will regularly update the links to the top three trackers right here</sub> :arrow_down:

TrackerAUCTrackerPrecision
ProContEXT40.10ODTrack55.80
KeepTrack39.60KeepTrack55.50
ODTrack39.60ProContEXT54.50

:fire::fire::fire: <sub>We will regularly update the links to the top three trackers right here</sub> :arrow_up:

How to benchmark:

Download the dataset

You can download our dataset here.

Run on our dataset

Please follow these step-by-step instructions to evaluate your algorithm.

Evaluation on server

Challenge Phases: Please note that our challenge is divided into two phases: the public phase and the private phase.

We have implemented these limits to maintain a clean and readable leaderboard. We encourage all challengers to submit only their most competitive results on the public leaderboard. To evaluate your results, please follow this tutorial.

Annotation tool and meta information

Annotation tool and meta information can be found here.

Citation

If you find our work valuable, we kindly ask you to consider citing our paper and starring ⭐ our repository. Our implementation includes dataset and useful tools and we hope it make life easier for the VOT research community.

@inproceedings{liu2024ntvot,
  title={NT-VOT211: A Large-Scale Benchmark for Night-time Visual Object Tracking},
  author={Yu Liu and Arif Mahmood and Muhammad Haris Khan},
  booktitle={Proceedings of the Asian Conference on Computer Vision (ACCV)},
  pages={to be announced},
  year={2024},
  organization={Springer}
}

Acknowledgments

The dataloader code borrows heavily from PyTracking.

Maintenance

Please open a GitHub issue for any help. If you have any questions regarding the technical details, feel free to contact us.

License

MIT License