In the last years, a substantial body of work has focused on the development of cognitive approaches leveraging on a symbolic model representation of the objects belonging to a scene and their relations. Specifically in the domains of Video Surveillance and Ambient Intelligence, recent advances in image, multimedia and video analysis, as well as in object detection and tracking, activity and gesture recognition have highlighted the need to move away from the traditional pattern matching and statistical approaches. Hybrid approaches that couple mining, learning and statistical reasoning with knowledge-driven solutions have emerged that support the development of more flexible, meaningful and robust solutions for improving the analytics aspects. Under this paradigm, Semantic Web technologies have attracted growing interest, bringing to the table the ability to formally capture intended semantics and to support automated reasoning, integration, management and sharing of information. Ontologies provide the means for formally capturing domain information, whereas the underpinning semantics formalise context-aware interpretation services that enable the effective coupling of statistical and logic-based reasoning, making available new forms of business intelligence and video analytics. However, the potential of Semantic Web technologies in addressing key requirements is yet to be fully explored: temporal reasoning, scalability, uncertainty handling, reasoning with noisy and missing information, as well as the theoretical aspects that underpin the combination of Semantic Web technologies with other fields of research, such as open and closed-world assumption, impose new challenges.

 

The workshop aims to favour the meeting of the Semantic Web, video and image analysis scholars and practitioners for facilitating the adoption of Semantic Web technologies in the domains of Video Surveillance and Ambient Intelligence. With this workshop we want to pave the way towards the proposition of novel hybrid approaches where results produced by traditional bottom-up low-level feature-based vision algorithms are fused with top-down approaches based on Semantic Web technologies.

 

We encourage original papers from scholars and practitioners aiming at defining, developing and demonstrating use of Semantic Web and knowledge-driven technologies addressing the, but not limited to, following research issues:

 

- Object detection and tracking;

- Event modelling and detection;

- Ontology-based recognition of activities, actions and gestures;

- Context-aware behaviour analysis;

- Hybrid frameworks for activity detection;

- People tracking and counting in crowd environment;

- Trajectories analysis and retrieval;

- Semantically-enriched situation awareness in JDL;

- Uncertainty in ontology-based systems for video analysis;

- Activity recognition under uncertainty, noise and incomplete data 

- Ontology design patterns for video surveillance and AmI;

- Description Logics and abductive reasoning for video surveillance and AmI;

- Combination of ontologies and rules for video and image analysis;

- Rule-based engine performance evaluation on video analysis datasets;

- SPARQL and rule engines for Video Analytics;

- RDF stream reasoning and video analysis;

- Coupling of Semantic Complex Event Processing with Machine Learning for event detection;

 

We seek full technical research papers with a length of up to 12 pages. All papers must represent original and unpublished work that is not currently under review. Papers will be evaluated according to their significance, originality, technical content, style, clarity, and relevance to the workshop. At least one author of each accepted paper is expected to attend the workshop (workshop attendees must pay the ISWC workshop registration fee as well as the conference registration fee). Submissions must adhere to the Springer LNCS format and be made through the EasyChair conference system (link will be available soon).

 

Accepted papers will be included in the workshop proceedings and will be published on the workshop web page.

 

A selection of best papers will be published in a special issue on Journal of Ambient Intelligence and Humanized Computing (Springer).

 

Pierluigi Ritrovato - Department of Information Engineering, Electrical Engineering and Applied Mathematics, Lab of Intelligent Machine for Video, Audio and Image Analysis (MIVIA), University of Salerno, Italy (pritrovato@unisa.it)

 

Georgios Meditskos - Information Technologies Institute, Centre for Research and Technology - Hellas, Greece (gmeditsk@iti.gr)

 

Grigoris Antoniou, University of Huddersfield, UK

Antonis Bikakis, University College London, UK

Luc Brun, ENSICAEN and GreyC Laboratories, France 

Marco Cristani, University of Verona, Italy 

Stamatia Dasiopoulou, Pompeu Fabra University, Spain

Pasquale Foggia, University of Salerno, Italy

Xiaoyi Jiang, University of Munster, Germany

Ioannis Kompatsiaris, Centre for Research and Technology - Hellas, Greece 

Efstratios Kontopoulos, Centre for Research and Technology - Hellas, Greece 

Brian Lovell, The Queensland University, Australia

Jose M. Martinez Sanchez, Universidad Autonoma de Madrid, Spain

Jose Manuel Molina Lopez, Universidad Carlos III de Madrid, Spain 

Nicolai Petrov, University of Groningen, Netherlands

Daniele Riboni, University of Milan, Italy

Giorgos Stamou, National Technical University of Athens, Greece

Giorgos Stoilos, National Technical University of Athens, Greece

Mario Vento, University of Salerno, Italy

Juan Ye, University of St. Andrews, UK