Research
Research - Demos
Segmentation is based on mean signal amplitude distribution, whereas classification utilizes an additional characteristic related to the frequency. The classification algorithm may be used either in conjunction with the segmentation algorithm, in which case it verifies or refutes a music-speech or speech-music change, or autonomously, with given audio segments.
Lifelike animal motion using as input captured video sequences of animal movement and prior anatomy knowledge.
We present a method for a 3D snake model construction and terrestrial snake locomotion synthesis in 3D virtual environments using image sequences.
This work is focused to propose an framework for synthesizing novel, arbitrarily long animations of periodic dances.
We define and analyze the general problem of partitioning a continuous curve into N parts with equal chords. The goal is to locate N-1 consecutive curve points, so that the curve can be divided into N segments with equal length chords.
We also present two applications of the Curve Equipartition Problem on Polygonal Approximation and Video Summarization.
This work is focused to propose a framework for interactive image segmentation problem. The goal of interactive image segmentation is to classify the image pixels into foreground and background classes, when some foreground and the background markers are given.
This work is focused to propose an unsupervised texture image segmentation framework with unknown number of regions, which involves feature extraction and classification in feature space, followed by flooding and merging in spatial domain.
This work is focused to represent a given 2D shape with an automatically determined number of ellipses, so that the total area covered by the ellipses is equal to the area of the original shape without any assumption or prior knowledge about the object structure.
We present a novel solution to the problem of Unconstrained Polygonal Fitting of 2D Shapes.
We present a novel solution to the problem of segmenting and splitting images of cells in unsupervised manner.
We present the definition and a solution to the problem of unsupervised image sorting.
This work is focused to propose an framework automatic enhancement and identification of the geological faults, using bathymetric/elevation images.
This work is focused to propose an unsupervised isocontour based segmentation method, that is applied on the detection of topographic highs with arbitrary basal shapes on Digital Elevation Models (DEMs).
This work is focused on the development of a system that analyses human motion and recognizes automatically the human action and activity in sports using image sequences.
Given two action sequences, our method discovers all pairs of similar subsequences, i.e. subsequences that represent the same action.
We present a solution to the problem of discovering all temporally periodic segments of a given video.
This work is focused on finding the entire community structure of a network, based on local interactions between neighboring nodes and on an unsupervised distributed hierarchical clustering algorithm.
A flow propagation algorithm (FlowPro) that finds the community of a node in a complex network.
This work is focused to propose a framework for recommendations. The goal is to predict the ratings of users for specific items, based on an analysis of previous consumer preferences.
We present DTEC system: A Dual Training Error based Correction Approach applied on Recommender Systems
We propose a framework to identify anomalous rating profiles, where each attacker hurriedly creates profiles that inject into the system an unspecified combination of random ratings and specific ratings, without any prior knowledge of the existing ratings .
We present the approach, we used as team "DataLab HMU.GR", for the ACM RecSys Challenge 2022. The challenge aims to predict the item that was purchased for a given sequence of item views (session).
We propose awe propose an efficient deterministic method based on Expectation-Maximization (EM) to solve the challenging problem of the tourist trip design or Personalized Itinerary Recommendation (PIR) with POI categories.
We show that the ranking of a country changes significantly when we take into consideration its scientific personnel that has migrated abroad (i.e., brain-drain).