Need help with TinyBenchmark?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

382 Stars 59 Forks MIT License 62 Commits 8 Opened issues


Scale Match for Tiny Person Detection(WACV2020), Official link of the dataset

Services available


Need anything else?

Contributors list

# 263,439
48 commits
# 376,222
14 commits

Scale Match for Tiny Person Detection

[paper] [ECCVW] [challenge] [ECCVW sumarry]

The annotaions of test set have released aready !!!
Baidu Yun, password:pmcq and Google Driver

For how to use the test_set annotation to evaluate, please see Evaluation

TODO list

  1. add a tutorial that how to train on TinyPerson with scale match on COCO
  2. add a tutorial that how to train on other dataset
  3. add a tutorial that how to train a strong baseline for competetion


TinyPerson Dataset

The dataset will be used to for ECCV2020 workshop RLQ-TOD'20 @ ECCV, TOD challenge

Download link:

Official Site: recomended, download may faster
Baidu Pan password: pmcq
Google Driver

For more details about TinyPerson dataset, please see Dataset.

Tiny Citypersons

Baidu Pan password:vwq2

Tiny Benchmark

The benchmark is based on maskrcnn_benchmark and citypersons code.

For more details about the benchmark, please see Tiny Benchmark.

Scale Match


If you use the code and benchmark in your research, please cite:

  title={Scale Match for Tiny Person Detection},
  author={Yu, Xuehui and Gong, Yuqi and Jiang, Nan and Ye, Qixiang and Han, Zhenjun},
  booktitle={The IEEE Winter Conference on Applications of Computer Vision},
And if the ECCVW challenge sumarry do some help for your research, please cite:
  title={The 1st Tiny Object Detection Challenge: Methods and Results},
  author={Yu, Xuehui and Han, Zhenjun and Gong, Yuqi and Jan, Nan and Zhao, Jian and Ye, Qixiang and Chen, Jie and Feng, Yuan and Zhang, Bin and Wang, Xiaodi and others},
  journal={arXiv preprint arXiv:2009.07506},

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.