HANDWRITTEN DIGITS RECONSTRUCTION FROM UNLABELLED EMBEDDINGS

Thomas Thebaud (1) , Gaël Le Lan (1) , Anthony Larcher (2)

1 Orange Labs [Cesson-Sévigné]
2 LIUM - Laboratoire d'Informatique de l'Université du Mans

In this paper, we investigate template reconstruction attack of touchscreen biometrics, based on handwritten digits writer verification. In the event of a template database theft, we show that reconstructing the original drawn digit from the embeddings is possible without access to the original embedding encoder. Using an external labelled dataset, an attack encoder is trained along with a Mixture Density Recurrent Neural Network decoder. Thanks to an alignment flow, initialized with Linear Discriminant Analysis and Procrustes, the transfer function between the output space of the original and the attack encoder is estimated. The successive application of transfer function and decoder to the stolen embeddings allows to reconstruct the original drawings, which can be used to spoof the behavioural biometrics system.

Mots clés: Alignment flow; Mixture Density Network; Behavioral biometrics; Template reconstruction attack

Domaines: Intelligence artificielle [cs.AI]

Thomas Thebaud, Gaël Le Lan, Anthony Larcher. HANDWRITTEN DIGITS RECONSTRUCTION FROM UNLABELLED EMBEDDINGS. ICASSP, Jun 2021, Toronto, Canada. ⟨hal-03262968⟩