shuida/lasot_evaluation_toolkit

Evaluation of trackers on a large-scale benchmark LaSOT.

LaSOT_Evaluation_Toolkit

This toolkit is utilized for evaluating trackers' performance on a large-scale benchmark LaSOT (https://cis.temple.edu/lasot/).

Notification

There is a problem with the data sever. Please use the following links to download dataset:

• Download the whole dataset through Google driver: https://bit.ly/LaSOTAll
• Download each category through Google driver: https://bit.ly/LaSOTEach
• Download the whole dataset through Baidu Pan: https://pan.baidu.com/s/1UbcQIU-Fpps7Jqq4WHRRkA
• Download each category through Baidu Pan: https://pan.baidu.com/s/1xFANiqkBHytE7stMOLUpLQ

Usage

• Download the repository, unzip it to your computer
• Download tracking result, unzip it to folder tracking_results/
• Run run_tracker_performance_evaluation.m in Matlab

Notes

In the file run_tracker_performance_evaluation.m, you can

• change evaluation_dataset_type (line 25) for evaluation on all 1,400 sequences or 280 testing sequences
• change norm_dst (line 28) for precision or normalized precision plots

In the file utils/plot_draw_save.m

• change the plotting settings to get the appropriate plots

Citation

If you use LaSOT and this evaluation toolkit for you researches, please consider citing our paper:

• H. Fan*, L. Lin*, F. Yang*, P. Chu*, G. Deng, S. Yu, H. Bai, Y. Xu, C. Liao, and H. Ling. LaSOT: A High-quality Benchmark for Large-scale Single Object Tracking. In CVPR, 2019.

Contact

If you have any questions on LaSOT, please feel free to contain Heng Fan at hengfan@temple.