Ioannis Pitas

TL;DR: Two fingerprinting approaches are reviewed in this paper: an image fingerprinting technique that makes use of color-based descriptors, R-trees and linear discriminant analysis (LDA) and a video fingerprinting method that utilizes information about the appearances of actors in videos along with an efficient search strategy.

TL;DR: A new watermarking algorithm for still images that is robust to compression, filtering, cropping, translation, rotation and scaling, and has a self-similar structure that accelerates the detection procedure.

TL;DR: A method for accurate and computationally efficient registration of 3-D shapes including curves and surfaces based on the iterative closest point (ICP) algorithm using Morphological Voronoi tessellation method to construct the Vor onoi regions around the seed points with respect to a certain distance metric.

TL;DR: The proposed K-A3 introduces novel optimization criteria to standard adversarial attack methodologies, inspired by the K-Anonymity principles, and the proposed M-SVDD-D can be used to prevent adversarial attacks in black-box attack settings.

TL;DR: The proposed Single-hidden Layer Feedforward Neural networks approach is evaluated by using the state-of-the-art Bag of Features-based action video representation on three publicly available action recognition databases, where it outperforms two commonly used video representation combination approaches, as well as the best single-descriptor classification outcome.