BEE24 Dataset

Multi-Object Tracking MOT Benchmark IEEE TIP 2025 Small Objects

36 Videos

446K Annotations

8+ Object Types

TIP Published In

About

BEE24 includes a total of 36 videos and 446,908 annotations. BEE24 challenges models to track multiple similar-appearing small objects with complex motions over long periods. It serves as a testbed for MOT research in more complex and diverse real-world scenes, significantly contributing to tracking general objects, including pedestrians, vehicles, drones, animals, insects, etc.

The dataset is designed to expose limitations of existing trackers under complex motion patterns, occlusion, and appearance similarity, pushing the boundary of multi-object tracking research in unconstrained environments.

The leaderboard of existing trackers, as well as training and testing sequences, can be found in the code repository and the paper.

News

(2025.01) Our paper is online! Please see IEEE TIP.
(2024.12) BEE24 is accepted by IEEE TIP! 🎊🎊🎊
(2024.10) We have released the BEE24 dataset!

Download

The dataset is publicly available from the following sources. All versions are identical in content.

Google Drive no password

References

If you find this dataset useful, please cite:

@ARTICLE{10851814, author  = {Cao, Xiaoyan and Zheng, Yiyao and Yao, Yao and Qin, Huapeng and Cao, Xiaoyu and Guo, Shihui}, journal = {IEEE Transactions on Image Processing}, title   = {TOPIC: A Parallel Association Paradigm for Multi-Object Tracking Under Complex Motions and Diverse Scenes}, year    = {2025}, volume  = {34}, pages   = {743-758}, doi     = {10.1109/TIP.2025.3526066} }

Contact

For questions or suggestions, contact us at caoxiaoyan@stu.pku.edu.cn.