KAIST Multispectral Pedestrian Detection Benchmark
2021-05-10 16:31:42
<< KAIST Multispectral Pedestrian Detection Benchmark >>
Download links for requested dataset
[All, Annotation files (.txt, 14.3MB)]
Download links for part of our dataset
Train
Set 00 / Day / Campus / 5.92GB / 17,498 frames [jpeg images]
Set 01 / Day / Road / 2.82GB / 8,035 frames [jpeg images]
Set 02 / Day / Downtown / 3.08GB / 7,866 frames [jpeg images]
Set 03 / Night / Campus / 2.40GB / 6,668 frames [jpeg images]
Set 04 / Night / Road / 2.88GB / 7,200 frames [jpeg images]
Set 05 / Night / Downtown / 1.01GB / 2,920 frames [jpeg images]
Test
Set 06 / Day / Campus / 4.78GB / 12,988 frames [jpeg images]
Set 07 / Day / Road / 3.04GB / 8,141 frames [jpeg images]
Set 08 / Day / Downtown / 3.50GB / 8,050 frames [jpeg images]
Set 09 / Night / Campus / 1.38GB / 3,500 frames [jpeg images]
Set 10 / Night / Road / 3.75GB / 8,902 frames [jpeg images]
Set 11 / Night / Downtown / 1.33GB / 3,560 frames [jpeg images]
Toolbox
To run multispectral acf algorithm, the modified Piotr's Computer Vision Toolbox is required.(Original toolbox)
Modified toolbox - compatible with 4ch inputs
* Please note that you must modify directory structure to run 'detector/acfDemoKAIST.m'.
Citation
If you use our dataset for your research, please consider citing our paper.
@inproceedings{CVPR15MultispectralPed,
author = {Soonmin Hwang and Jaesik Park and Namil Kim and Yukyung Choi and In So Kweon},
title = {Multispectral Pedestrian Detection: Benchmark Dataset and Baseline},
booktitle = {CVPR},
year = {2015},
}
Thank you for your attention.