Decription :
Multi-modal biometric synchronized sensing dataset (MM-BioSync) is a database which incorporates behavioral biometric modalities (motion sensors, touchscreen and keystroke dynamics) alongside front-face videos, collected in a synchronized manner using android-based smartphone devices, for the purpose of continuous and active biometric verification.
The need for multi-modal dataset designed specifically for active biometric verification on smartphones is critical. Moreover, current datasets don’t provide enough modalities with synchronization. To tackle this problem, we introduce a new multi-modal biometric dataset, with various behavioral modalities synchronized with front-face camera, to enhance the process of active verification.
Acquisition details:
The MM-BioSync dataset is composed of 50 subjects, each subject performed two different tasks on an Android app: reading a text (reading task) and answering questions about the text (writing task). For each subject, five reading and five writing sessions are collected, with 120 seconds of image frames at the beginning of each session. In total there are 500 sessions of data included in the MM-BioSync dataset. Below are shown some samples of the MM-Biosync dataset for the different tasks of reading and writing:
References:
Any publication using this dataset must cite the following paper:…
Download:
The link to download the dataset will be provided after receiving a signed license
agreement. A scanned copy of the license agreement must be sent by email at: mm_biosync@g.enp.edu.dz.
Any further information or questions regarding the MM-Biosync dataset can be addressed to :
- Youcef OUADJER
- youcef.ouadjer@g.enp.edu.dz
- Mourad ADNANE
- mourad.adnane@g.enp.edu.dz