A5084122016- Advanced Image and Video Retrieval Techniques
- Image Retrieval and Classification Techniques
- Video Analysis and Summarization
- Semantic Web and Ontologies
- Complex Network Analysis Techniques
- EEG and Brain-Computer Interfaces
- Context-Aware Activity Recognition Systems
- Multimodal Machine Learning Applications
- Human Pose and Action Recognition
- Anomaly Detection Techniques and Applications
- Web Data Mining and Analysis
- Video Surveillance and Tracking Methods
- Music and Audio Processing
- Data Management and Algorithms
- Digital Media Forensic Detection
- Advanced Vision and Imaging
- Advanced Text Analysis Techniques
- Biomedical Text Mining and Ontologies
- Recommender Systems and Techniques
- Dementia and Cognitive Impairment Research
- Spam and Phishing Detection
- Misinformation and Its Impacts
- Neural dynamics and brain function
- Topic Modeling
- Remote-Sensing Image Classification
Centre for Research and Technology Hellas
2016-2025
Information Technologies Institute
2016-2025
China Philanthropy Research Institute
2011-2023
Information Technology Institute
2012-2021
Aristotle University of Thessaloniki
1998-2021
Hella (Germany)
2008-2020
Universität Ulm
2019
Fraunhofer Institute of Optronics, System Technologies and Image Exploitation
2018
Brigham and Women's Hospital
2016
Harvard University
2016
Abstract Artificial Intelligence (AI)‐based systems are widely employed nowadays to make decisions that have far‐reaching impact on individuals and society. Their might affect everyone, everywhere, anytime, entailing concerns about potential human rights issues. Therefore, it is necessary move beyond traditional AI algorithms optimized for predictive performance embed ethical legal principles in their design, training, deployment ensure social good while still benefiting from the huge of...
Online social and news media generate rich timely information about real-world events of all kinds. However, the huge amount data available, along with breadth user base, requires a substantial effort filtering to successfully drill down relevant topics events. Trending topic detection is therefore fundamental building block monitor summarize originating from sources. There are wide variety methods variables they greatly affect quality results. We compare six on three Twitter datasets...
Oil spill is considered one of the main threats to marine and coastal environments. Efficient monitoring early identification oil slicks are vital for corresponding authorities react expediently, confine environmental pollution avoid further damage. Synthetic aperture radar (SAR) sensors commonly used this objective due their capability operating efficiently regardless weather illumination conditions. Black spots probably related spills can be clearly captured by SAR sensors, yet...
An approach to knowledge-assisted semantic video object detection based on a multimedia ontology infrastructure is presented. Semantic concepts in the context of examined domain are defined an ontology, enriched with qualitative attributes (e.g., color homogeneity), low-level features model components distribution), spatial relations, and processing methods clustering). Web technologies used for knowledge representation RDF(S) metadata standard. Rules F-logic describe how tools analysis...
In this paper, a novel approach to video temporal decomposition into semantic units, termed scenes, is presented. contrast previous segmentation approaches that employ mostly low-level visual or audiovisual features, we introduce technique jointly exploits and high-level features automatically extracted from the auditory channel. This built upon well-known method of scene transition graph (STG), first by introducing new STG approximation reduced computational cost, then extending unimodal...
An image analysis scheme can automate the detection of landmarks and events in large collections, significantly improving content-consumption experience.
This paper deals with content-based large-scale image retrieval using the state-of-the-art framework of VLAD and Product Quantization proposed by Jegou as a starting point. Demonstrating an excellent accuracy-efficiency trade-off, this has attracted increased attention from community numerous extensions have been proposed. In work, we make in-depth analysis that aims at increasing our understanding its different processing steps boosting overall performance. Our involves evaluation (both...
This paper focuses on detecting and localizing anomalous events in videos of crowded scenes, i.e., divergences from a dominant pattern. Both motion appearance information are considered, so as to robustly distinguish different kinds anomalies, for wide range scenarios. A newly introduced concept based swarm theory, histograms oriented swarms (HOS), is applied capture the dynamics environments. HOS, together with well-known gradients, combined build descriptor that effectively characterizes...
Location extraction, also called “toponym extraction,” is a field covering geoparsing, extracting spatial representations from location mentions in text, and geotagging, assigning coordinates to content items. This article evaluates five “best-of-class” extraction algorithms. We develop geoparsing algorithm using an OpenStreetMap database, geotagging language model constructed social media tags multiple gazetteers. Third-party work evaluated includes DBpedia-based entity recognition...
Machine learning bias and fairness have recently emerged as key issues due to the pervasive deployment of data-driven decision making in a variety sectors services. It has often been argued that unfair classifications can be attributed training data, but previous attempts 'repair' data led limited success. To circumvent shortcomings prevalent repairing approaches, such those weight samples sensitive group (e.g. gender, race, financial status) based on their misclassification error, we...
Background: Assistive technology, in the form of a smart home environment, is employed to support people with dementia. Objectives: To propose system for continuous and objective remote monitoring problematic daily living activity areas design personalized interventions based on fee dback clinical observations improving cognitive function health-related quality life. Methods: The assistive technology proposed system, including wearable, sleep, object motion, presence, utility usage sensors,...