A5059708050- Advanced Vision and Imaging
- Robotics and Sensor-Based Localization
- Human Pose and Action Recognition
- Computer Graphics and Visualization Techniques
- Advanced Image and Video Retrieval Techniques
- Virtual Reality Applications and Impacts
- 3D Shape Modeling and Analysis
- Image and Video Quality Assessment
- Advanced Image Processing Techniques
- Image Enhancement Techniques
- Image Retrieval and Classification Techniques
- Augmented Reality Applications
- Medical Image Segmentation Techniques
- Video Surveillance and Tracking Methods
- 3D Surveying and Cultural Heritage
- Optical measurement and interference techniques
- Video Analysis and Summarization
- Robot Manipulation and Learning
- Hand Gesture Recognition Systems
- Video Coding and Compression Technologies
- Advanced Neural Network Applications
- Image Processing Techniques and Applications
- Generative Adversarial Networks and Image Synthesis
- UAV Applications and Optimization
- COVID-19 diagnosis using AI
Information Technologies Institute
2016-2025
Centre for Research and Technology Hellas
2016-2025
China Philanthropy Research Institute
2020-2023
Universidad Politécnica de Madrid
2019
Hella (Germany)
2018
Information Technology Institute
2012-2015
Aristotle University of Thessaloniki
2004-2013
Pennsylvania State University
2006
The problem of robust, realistic and especially fast 3-D reconstruction objects, although extensively studied, is still a challenging research task. Most the state-of-the-art approaches that target real-time applications, such as immersive reality, address mainly synthesizing intermediate views for given view-points, rather than generating single complete surface. In this paper, we present multiple-Kinect capturing system novel methodology creation accurate, realistic, full reconstructions...
Adopting effective techniques to automatically detect and identify small drones is a very compelling need for number of different stakeholders in both the public private sectors. This work presents three original approaches that competed grand challenge on “Drone vs. Bird” detection problem. The goal one or more appearing at some time point video sequences where birds other distractor objects may be also present, together with motion background foreground. Algorithms should raise an alarm...
Learning based approaches for depth perception are limited by the availability of clean training data. This has led to utilization view synthesis as an indirect objective learning estimation using efficient data acquisition procedures. Nonetheless, most research focuses on pinhole monocular vision, with scarce works presenting results omnidirectional input. In this work, we explore spherical 360 in a self-supervised manner and demonstrate its feasibility. Under purely geometrically derived...
The focus of research into 5G networks to date has been largely on the required advances in network architectures, technologies, and infrastructures. Less effort put applications services that will make use exploit flexibility built upon concept software-defined networking (SDN) function virtualization (NFV). Media-based are amongst most demanding services, requiring large bandwidths for high audio-visual quality, low-latency interactivity, sufficient infrastructure resources deliver...
Human motion estimation is a topic receiving high attention during the last decades. There vast range of applications that employ human tracking, while industry continuously offering novel tracking systems, which are opening new paths compared to traditionally used passive cameras. Motion algorithms, in their general form, estimate skeletal structure body and consider it as set joints limbs. However, systems usually work on single sensor basis, hypothesizing occluded parts. We hereby present...
The latest developments in 3D capturing, processing, and rendering provide means to unlock novel application pathways. main elements of an integrated platform, which target tele-immersion future applications, are described this paper, addressing the tasks real-time robust human shape/appearance reconstruction, skeleton-based motion tracking. More specifically, initially, details a multiple RGB-depth (RGB-D) capturing system given, along with sensors' calibration method. A robust, fast...
This paper presents the second edition of "drone-vs-bird" detection challenge, launched within activities 16-th IEEE International Conference on Advanced Video and Signal-based Surveillance (AVSS). The challenge's goal is to detect one or more drones appearing at some point in video sequences where birds may be also present, together with motion background foreground. Submitted algorithms should raise an alarm provide a position estimate only when drone while not issuing alarms birds, nor...
This paper presents the 4-th edition of "drone-vs-bird" detection challenge, launched in conjunction with 17-th IEEE International Conference on Advanced Video and Signal-based Surveillance (AVSS). The objective challenge is to tackle problem detecting presence one or more drones video scenes where birds may suddenly appear, taking into account some important effects such as background foreground motion. proposed solutions should identify localize scene only when they are actually present,...
We demonstrated a 3D holoscopic video system for 3DTV application. showed that using field lens and square aperture significantly reduces the vignetting problem associated with relay achieves over 95 percent fill factor. The main such is nonlinear distortion during image capturing, which can seriously affect reconstruction process display. mainly includes radial (intrinsic) microlens array perspective (extrinsic). This task of future work. Our results also show SS coding approach performs...
The popularity of Unmanned Aerial Vehicles (UAVs) is increasing year by and reportedly their applications hold great shares in global technology market. Yet, since UAVs can be also used for illegal actions, this raises various security issues that needs to encountered. Towards end, UAV detection systems have emerged detect further anticipate inimical drones. A very significant factor the maximum range which system's senses "see" an upcoming UAV. For those employ optical cameras detecting...
We introduce HUMAN4D, a large and multimodal 4D dataset that contains variety of human activities simultaneously captured by professional marker-based MoCap, volumetric capture an audio recording system. By capturing 2 female $2$ male actors performing various full-body movements expressions, HUMAN4D provides diverse set motions poses encountered as part single- multi-person daily, physical social (jumping, dancing, etc.), along with multi-RGBD (mRGBD), data. Despite the existence multi-view...
Pano3D is a new benchmark for depth estimation from spherical panoramas. It aims to assess performance across all traits, the primary direct targeting precision and accuracy, also secondary boundary preservation smoothness. Moreover, moves beyond typical intra-dataset evaluation inter-dataset assessment. By disentangling capacity generalize in unseen data into different test splits, represents holistic 360 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"...
This paper presents a novel methodology for content-based search and retrieval of 3D objects. After proper positioning the objects using translation scaling, set functionals is applied to model producing new domain concentric spheres. In this domain, applied, resulting in descriptor vector which completely rotation invariant thus suitable matching. Further, weights are assigned each descriptor, so as significantly improve results. Experiments on two different databases performed evaluate...
Measuring the similarity between three-dimensional (3-D) objects is a challenging problem, with applications in computer vision, molecular biology, graphics, and many other areas. This paper describes novel method for 3-D model content-based search based on Generalized Radon Transform querying by-3-D-model approach. A set of descriptor vectors extracted using Radial Integration (RIT) Spherical (SIT), which represent significant shape characteristics. After proper alignment models, are...
Depth perception is considered an invaluable source of information for various vision tasks. However, depth maps acquired using consumer-level sensors still suffer from non-negligible noise. This fact has recently motivated researchers to exploit traditional filters, as well the deep learning paradigm, in order suppress aforementioned non-uniform noise, while preserving geometric details. Despite effort, denoising open challenge mainly due lack clean data that could be used ground truth. In...
In this paper, a 3D shape-based approach is presented for the efficient search, retrieval, and classification of protein molecules. The method relies primarily on geometric structure proteins, which produced from corresponding PDB files secondarily their primary secondary structure. After proper positioning structures, in terms translation scaling, Spherical Trace Transform applied to them so as produce geometry-based descriptor vectors, are completely rotation invariant perfectly describe...
Recent advances in media capture and processing technologies have enabled new forms of true 3-D content that increase the degree user immersion. The demand for more engaging entertainment means distributors broadcasters need to fine-tune their delivery mechanisms over Internet as well develop models quantifying predicting experience these content. In work described this paper, we undertake one first studies into quality (QoE) real-time streamed virtual reality (VR) headsets purposes, context...
In this paper, a novel skeleton-based approach to human time-varying mesh (H-TVM) compression is presented. The topic of TVM new and has many challenges, such as handling the lack obvious mapping vertices across frames variable connectivity frames, while maintaining efficiency, which are most important ones. Very few works exist in literature, not all challenges have been addressed yet. addition, developing an efficient real-time solution, above, obviously difficult task. We attempt address...