SCUT-DHGA-br: A Synthetic Database for Dynamic Hand Gesture Authentication
SCUT-DHGA-br (SD-br) dataset is a syntheticdataset derived from the SCUT-DHGA (SD) dataset specifically for ro-bustness evaluation. It replaces the backgrounds of eachvideo in the SD dataset with 3627 complex backgrounds collected from the Internet and the real world, such asairports, classrooms, malls, museums, subways, etc. Thebrightness of the hand area is also adjusted by a randombrightness factor to mimic the diverse illumination inpractical scenarios. The size of the SD-br dataset is thesame as the SD dataset. It is collected by BIP Lab, School of Automation Science and Engineering, South China University of Technology(SCUT). Any one or group is allowed to use this database free of charge.
Images are labeled as follow: train/test_session_ID_gesture_sample.bmp,
where “train/test” indicates whether the video sample belongs to the training set or the testing set ,
“session” stands for session number which can be "1" or "2"
"ID" stands for client's ID,
“gesture” ranges from 1 to 6 standing for six gesture types,
“sample” stands for the repetitions of the sample, ranging from 1 to 10,
The naming convention of SCUT-DHGA-br is same as the SCUT-DHGA database. please refer to https://github.com/SCUT-BIP-Lab/SCUT-DHGA for details.
The SCUT-DHGA-br is publicly available (free of charge) to the research community.
Unfortunately, due to privacy reasons, we cannot provide the database for commercial use.
Those interested in obtaining SCUT-DHGA-br should download release agreement, and send by email one signed and scanned copy to [email protected].
While reporting results using the SCUT-DHGA-br, please cite the following article:
@ARTICLE{10654331,
author={Zhang, Yufeng and Kang, Wenxiong and Song, Wenwei},
journal={IEEE Transactions on Information Forensics and Security},
title={Robust and Accurate Hand Gesture Authentication With Cross-Modality Local-Global Behavior Analysis},
year={2024},
volume={19},
number={},
pages={8630-8643},
keywords={Authentication;Videos;Feature extraction;Physiology;Robustness;Lighting;Spatiotemporal phenomena;Biometrics;hand gesture authentication;multimodal fusion;spatiotemporal analysis;behavioral characteristic representation},
doi={10.1109/TIFS.2024.3451367}}
Zhang Yufeng
Biometrics and Intelligence Perception Lab.
College of Automation Science and Engineering
South China University of Technology
Wushan RD.,Tianhe District,Guangzhou,P.R.China,510641
[email protected]
Song Wenwei
Biometrics and Intelligence Perception Lab.
College of Automation Science and Engineering
South China University of Technology
Wushan RD.,Tianhe District,Guangzhou,P.R.China,510641
[email protected]
Prof. Kang Wenxiong
Biometrics and Intelligence Perception Lab.
College of Automation Science and Engineering
South China University of Technology
Wushan RD.,Tianhe District,Guangzhou,P.R.China,510641
[email protected]