Leveraging Expert Models for Training Deep Neural Networks in Scarce Data Domains: Application to Offline Handwritten Signature Verification

no code implementations • 2 Aug 2023 • Dimitrios Tsourounis, Ilias Theodorakopoulos, Elias N. Zois, George Economou

This paper introduces a novel approach to leverage the knowledge of existing expert models for training new Convolutional Neural Networks, on domains where task-specific data are limited or unavailable.

Knowledge Distillation

A comprehensive study of sparse representation techniques for offline signature verification

no code implementations • 13 Jul 2018 • Elias N. Zois, Dimitrios Tsourounis, Ilias Theodorakopoulos, Anastasios Kesidis, George Economou

In this work, a feature extraction method for offline signature verification is presented that harnesses the power of sparse representation in order to deliver state-of-the-art verification performance in several signature datasets like CEDAR, MCYT-75, GPDS and UTSIG.

Attention-Aware Generative Adversarial Networks (ATA-GANs)

no code implementations • 25 Feb 2018 • Dimitris Kastaniotis, Ioanna Ntinou, Dimitrios Tsourounis, George Economou, Spiros Fotopoulos

Using the attention maps produced by a Teacher- Network we are able to improve the quality of the generated images as well as perform weakly object localization on the generated images.

object-detection Object Detection +2