Back to datasets
Dataset assetOpen Source CommunityComputer VisionNighttime Visual Tracking

NT-VOT211

NT‑VOT211 is a large‑scale night‑time visual object tracking benchmark created by Xinjiang University and other institutions. The dataset contains 211 diverse videos, totaling 211,000 meticulously annotated frames, covering eight attributes, and is designed to evaluate tracking algorithms under night‑time conditions. It was built to address the lack of suitable night‑time data in existing benchmarks. NT‑VOT211 finds applications in security surveillance, autonomous driving, and wildlife protection, especially for object tracking in low‑light environments.

Source
arXiv
Created
Oct 27, 2024
Updated
Oct 27, 2024
Signals
205 views
Availability
Linked source ready
Overview

Dataset description and usage context

NV‑VOT211 Dataset Overview

Summary

NV‑VOT211 is a newly proposed benchmark for evaluating nighttime visual object tracking algorithms. The dataset contains 211 diverse videos, providing 211,000 well‑annotated frames with eight attributes, including camera motion, deformation, fast motion, motion blur, small objects, interference, occlusion, and out‑of‑view.

Dataset Download

The dataset can be downloaded from the following link: NV‑VOT211 Dataset Download.

Evaluation Procedure

Running the Evaluation Script

Follow these steps to evaluate your algorithm: Evaluation Script Instructions.

Server Evaluation

The challenge is split into two phases: public and private.

  • Private Phase: Up to 100 submissions per day.
  • Public Phase: Up to 1 submission per month.

Evaluation results are detailed in this tutorial: Server Evaluation Guide.

Annotation Tools and Metadata

The annotation tools and related metadata are available here: Annotation Tools & Metadata.

Citation

If you find our work valuable, please cite our paper and star the repository.

bibtex
@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}
}
Need downstream help?

Pair the dataset with AI analysis and content workflows.

Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.

Explore AI studio