I
Ioannis Pitas
Researcher at Aristotle University of Thessaloniki
Publications - 826
Citations - 26338
Ioannis Pitas is an academic researcher from Aristotle University of Thessaloniki. The author has contributed to research in topics: Facial recognition system & Digital watermarking. The author has an hindex of 76, co-authored 795 publications receiving 24787 citations. Previous affiliations of Ioannis Pitas include University of Bristol & University of York.
Papers
More filters
Proceedings ArticleDOI
Optimal Multidimensional Cyclic Convolution Algorithms For Deep Learning And Computer Vision Applications
TL;DR: In this paper, the authors presented an optimal n-D cyclic convolution algorithm with minimal multiplicative complexity that is much faster than any competing convolutional algorithm internationally and methods for speeding up such optimal convolution algorithms on GPUs and multicore CPUs.
Proceedings ArticleDOI
An information theoretic approach to joint probabilistic face detection and tracking
TL;DR: A joint probabilistic face detection and tracking algorithm for combining a likelihood estimation and a prior probability estimation based on the fusion of an information theoretic tracking cue and a Gaussian temporal model is proposed.
Proceedings ArticleDOI
Face authentication based on morphological shape decomposition
TL;DR: First experimental results collected are very encouraging and indicate that the proposed method outperforms the (standard) dynamic link matching that is based on Gabor wavelets.
Proceedings ArticleDOI
Introducing the select and split fuzzy cell Hough transform
Vassilios Chatzis,Ioannis Pitas +1 more
TL;DR: A new variation of Hough transform that is recursively roughly split into fuzzy cells which are defined as fuzzy numbers can be used to detect contours in an image, with better accuracy, especially in noisy images.
Proceedings ArticleDOI
Two-dimensional vector median filters on mesh connected computers
G. Angelopoulos,Ioannis Pitas +1 more
TL;DR: A parallel algorithm for the calculation of vector median filters on mesh connected computers by taking into account the fact that adjacent filter windows share common input samples, and the computation of the common norms of adjacent windows could be done only once.