CVSP Group

Computer Vision, Speech Communication & Signal Processing

National Technical University of Athens

About The Group

Our goals are to advance and sustain basic research in both theoretical and application problems whose fundamentals lie in the intersection of various fields of engineering and applied sciences.

We bridge the gap between Signal Processing, Systems Theory, and Informatics to solve complex challenges in the modern digital world.

Core Research Domains

Computer Vision

Image understanding, object detection, 3D reconstruction, and visual scene analysis.

Audio & Speech Processing

Modeling of speech production, automatic speech recognition, and analysis of 3D acoustic scenes.

Machine Learning & AI

Cognitive systems, deep neural networks, and multimodal learning algorithms.

Robotics & Perception

Visual perception, autonomous navigation, and human-robot interaction (HRI).

Biomedicine

Advanced signal processing for medical imaging, diagnostics, and patient monitoring.

Nonlinear Systems

Mathematical morphology, fractals, chaos theory, and dynamical systems analysis.

Original Research

Child Robot Interaction

Child Robot Interaction

Multimodal Signal and Event Processing In Perception and Cognition

Multimodal Signal and Event Processing In Perception and Cognition

Graph Clustering for Image Segmentation

Graph Clustering for Image Segmentation

We propose a graph clustering approach for image segmentation by developing diffusion processes defined on arbitrary graphs.

Audiovisual Speech Inversion

Audiovisual Speech Inversion

We focus on recovering aspects of vocal tract’s geometry and dynamics from speech, a problem referred to as speech inversion.

Image Texture Modeling and Analysis

Image Texture Modeling and Analysis

We pursue a concise texture modeling, analysis and segmentation system for generic natural images.

Multigrid Geometric Active Contours

Multigrid Geometric Active Contours

We investigate multigrid techniques for the solution of the time-dependent PDEs of geometric active contour models in Computer Vision.