This system is able to recognize users from their writing style on a touchscreen.
The dataset has been gathered through the use of the touch-recorder android application
The performance of the system are very encouraging both in identification and verification:
| Italic | Block letters | 
|---|---|
![]()  | 
![]()  | 
![]()  | 
![]()  | 
![]()  | 
![]()  | 
Read more in the docs







