A5040972086- Image Retrieval and Classification Techniques
- Semantic Web and Ontologies
- Video Analysis and Summarization
- Advanced Image and Video Retrieval Techniques
- Natural Language Processing Techniques
- Biomedical Text Mining and Ontologies
- Topic Modeling
- Speech and dialogue systems
- Music and Audio Processing
- Remote-Sensing Image Classification
- Service-Oriented Architecture and Web Services
- Data Management and Algorithms
- Constraint Satisfaction and Optimization
- Geographic Information Systems Studies
- Context-Aware Activity Recognition Systems
- Advanced Database Systems and Queries
- Social Robot Interaction and HRI
- Medical Imaging Techniques and Applications
- Big Data and Business Intelligence
- Emotion and Mood Recognition
- Superconducting Materials and Applications
- Multimodal Machine Learning Applications
- Logic, Reasoning, and Knowledge
- Software Engineering Research
- Genomics and Phylogenetic Studies
Pompeu Fabra University
2014-2019
Information Technologies Institute
2004-2013
Information Technology Institute
2013
Centre for Research and Technology Hellas
2005-2011
Aristotle University of Thessaloniki
2005-2011
Universität Koblenz
2006
Koblenz University of Applied Sciences
2006
University of Bristol
2005
Institute of Informatics and Telematics
2004
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...
Knowledge representation and semantic annotation of multimedia documents typically have been pursued in two different directions. Previous approaches focused either on low-level descriptors, such as dominant colour, or the content dimension corresponding manual annotations, person vehicle. Here, a knowledge infrastructure an experimentation platform for to bridge directions are presented. Ontologies being extended enriched include audiovisual features descriptors. Additionally, tool that...
In this paper we describe SP-ACT, a hybrid framework for the derivation of high-level activity interpretations in context-aware environments, by defining combination OWL ontologies and SPARQL CONSTRUCT graph patterns. More specifically, native semantics is used to formally represent integrate activity-related information originated from different data sources, whereas (SPIN) rules further aggregate activities so as derive highlevel abstractions. The goal address limitations ontology-based...
Semantic analysis of multimedia data, such as images, image databases and videos, has attracted much attention in the past years. The extraction human centered descriptions, matching end users cognition, specifically detection identification events videos is a particularly challenging problem, due to volume diversity both low-level features that can be extracted from also corresponding high-level information conveyed. Numerous efforts have begun, attempting bridge semantic gap between data...
This paper surveys the approaches to video representation, focusing on semantic analysis for content-based indexing and retrieval. A problem of adaptive representation digital multimedia is critically assessed some novel ideas are presented. Furthermore, concept multimodality reevaluated redefined in order introduce modalities such as editing technique or affect audience.
In this paper we present an approach that combines multimedia reasoning and natural language processing for the semantic integration of automatic manual image annotations based on domain ontologies. We discuss how to apply transform descriptions queries into ontological representation allows users formulate formal semantics in intuitive manner, without need cope with complex structures unwieldy user interfaces. Illustrative experimental examples demonstrate added value.
In this paper, we report on adapting a Natural Language Generation system to Semantic Web datasets, and the results obtained in two triple verbalization challenges.
In this paper, we describe the principles and technologies that underpin development of an adaptive dialogue manager framework, tailored to carrying out human-agent conversations in a natural, robust flexible manner. Our research focus is twofold. First, investigation strategies can handle dynamically created user system actions, while still enabling agent adapt its actions various possibly changing contexts. Second, utilisation rich semantic annotations for capturing background knowledge,...
Efficient video content management and exploitation requires extraction of the underlying semantics, a non-trivial task associating low-level features image domain high-level semantic descriptions. In this paper, knowledge-assisted approach for extracting semantics domain-specific is presented. Domain knowledge considers both (color, motion, shape) spatial behavior (topological directional information). During preprocessing step, set over-segmented homogenous atom-regions generated their...