CARETAKER project

Ontology-Based Event Detection for Knowledge Discovery

Participants: Francois Bremond, Etienne Corvee, Jose Luis Patino Vilchis, Monique Thonnat.

The CARETAKER (Content Analysis and REtrieval Technologies to Apply Extraction to massive Recording) project aims at studying, developing and assessing multimedia knowledge-based content analysis, knowledge extraction components and metadata management sub-systems in the context of automated situation awareness, diagnosis and decision support. More precisely, CARETAKER will focus on the extraction of structured knowledge from large multimedia collections recorded over networks of cameras and microphones deployed in real sites. The produced audio-visual streams, if stored and automatically analyzed, represent a useful source of information in urban/environment planning, resource optimization, disabled/elderly person monitoring, etc., in addition to surveillance and safety issues.

We have considered two types of content knowledge: a first layer of primitive events that can be extracted from the raw data streams, such as ambient sounds, the degree of crowding present in the scene and the routes taken by individual people. A second layer of higher semantic events is defined based on more complex relationships between the primitive events and detected from longer term analysis.

Real testbed sites inside the metro of Roma and Torino, involving more than 30 sensors each (20 cameras and 10 microphones), has been provided. Additionally, the identification of the real user needs and beneficial use-case scenarios has served as a reference point for the correct framing of the semantic description scheme i.e. the ontology, the knowledge extraction components and the interface and demonstrator optimization.

Figure 1. Illustrations of Event Detection.

Within CARETAKER, the Orion project team is in charge of the long term tracking of objects of interest, the ontology-based event detection and of knowledge discovery.

• Ontology

  An ontology is the set of all concepts and relations between concepts shared by the community of a given domain. The ontology is particularly useful for experts of the application domain to use scene understanding systems in an autonomous way. In ORION, we have extended this year our video ontology toward two directions: (1) the ontology has been adapted for urban monitoring applications; (2) the video ontology has been combined with an ontology for audio events, to enrich the description of scenarios of interest.

• Tracking

  Tracking is one of the most studied topics in dynamic scene analysis. In ORION, as well as with other partners, novel algorithms available at the consortium have first been adapted to the specific environment of metro sites, and properly evaluated for the task. All the available context information is employed (e.g. rough calibration of the ground plane). In particular, one aim of CARETAKER is to increase robustness with respect to knowledge discovery: the detection and tracking of people and activity recognition should be performed on a long-term basis (more than one month).

• Event recognition and knowledge discovery

  The first objective in knowledge modeling and event recognition consists in developing new algorithms for the recognition of higher-level/composite events defined in the ontology from the user requirements, using the low-level primitive streams-of-data events coming from diverse sources of information (trajectories, audio/video activities), ontology-driven methods (i.e. scenario-based), or a mixture of both. For example, we are able to define and recognize when a luggage is being abandoned by a person based on two constraints: when a luggage remains in a predefined zone for a minimum predefined period of time and when it is located far enough to any passing persons, then this luggage is considered as 'abandoned luggage' as shown in figure 1. In this figure, the luggage labeled 'LUGGAGE 10' has been left apart by the person labeled 'PERSON 11' for a too long period of time and both person 11 and person 8 are located far enough from this luggage for it to be detected as abandoned.

  Figure 2. An abandoned luggage detected by the ontology-based event detector

  Figure 3. A jumping over barrier event detected in Roma subway, with the video.

  The second objective in knowledge modeling and event recognition is the investigation of unsupervised techniques for the recognition of both common events and unusual events. In all cases, the overall goal is to produce indexing information to feed the user system, for both the online supervision system and the offline retrieval one. Broadly speaking, different types of events exist, which can be characterized by their occurrence frequency, their expectedness, and their relevance. Their recognition poses different challenges that have been addressed using different models and methodologies. In ORION, the first results of primitive video events and object-of-interest trajectories are being analyzed using data mining tools. For instance, trajectories in a scene have been clustered into categories according to their entering and exiting a zone in the scene.

  Figure 3. Clustering of trajectories into 21 groups to match people activities in Torino subway (45min), 2052 trajectories.

  Figure 4. Analysis of the Hall zone: there is an increase of people after 6:45 .

For more information, see the CARETAKER Project Web-Site

 

CARETAKER - Content Analysis and REtrieval Technologies to Apply Knowledge Extraction to massive Recording

The CARETAKER project aimed at studying, developing and assessing multimedia knowledge-based content analysis, knowledge extraction components, and metadata management sub-systems in the context of automated situation awareness, diagnosis and decision support.

Impact

CARETAKER investigated techniques allowing the automatic extraction of relevant semantic metadata from raw multimedia, to explore the value of the extracted information to relevant users, and to demonstrate this in a framework that preserves the privacy of the individual. More precisely, CARETAKER focused on the extraction of a structured knowledge from multimedia collections recorded over a network of camera and microphones. Thus, the overall goal of CARETAKER was to investigate current and novel technologies to extract and exploit this information, by evaluating them in a real test case (Metro of Roma), while exploring the added-value of this technology for real users.

Main innovation

CARETAKER brought innovations in the following fields:

• Knowledge representation and acquisition: The explicit formal representation of the events, entities and relations in the scene is known as an 'ontology' allowing users to express their needs, define their scenario of interest or their potential queries to the knowledge system.
• Content extraction and knowledge discovery: These issues will be covered by the investigation of hybrid systems combining the advantages of data/event-driven probabilistic techniques and of user-driven ontology-based recognition techniques to model user knowledge, scenarios and entity relationships.
• Content access: Two innovative user-centred ontology- and data/event-driven demonstrators will be achieved providing web-service oriented access from low to high-level semantic events.

 

More details

• Mobility - The European Public Transport Magazine article: Caretaker puts knowledge to good use (June 2008) (PDF 1.486KB)
• Presentation (PDF 376KB)
• Flyer (PDF 570KB)
• A Tag-n-Track Module for the CARETAKER Project. Poster (PDF 1.566KB)
• On the Use of real-Time Agents in Distributed Video Analysis Systems. Poster (PDF 455KB)
• CARETAKER and the Multimedia Application Format (MAF) for Surveillance. Presentation (PDF 723KB)
• Annual Report 2008 (PDF 524KB)
• Annual Report 2007 (PDF 386KB)




• Annual Report 2006 (PDF 196KB)

Administrative Details

• CARETAKER (IST-4-027231) is a Specific Targeted Project of the European Union's 6th Framework Program - call 4.
• CARETAKER started on 1 March 2006 and finished on 30 August 2008, and the overall Budget is 2.9 million euros.
• Eight partners from seven European countries were involved in the project.

List of Participants

• Thales Communications (prime contractor), France
• Multitel, Belgium
• INRIA, France
• Kingston University, United Kingdom
• IDIAP, Switzerland
• ATAC, Italy
• Solid, Finland
• Brno University of Technology, Czech Republic

Contact Persons

Bertrand Ravera

Thales Communications

Multimedia Processing Laboratory

146 Boulevard de Valmy, BP 82, 92704 Colombes cedex, FRANCE

Tel: +33 1 46133172

Fax: +33 1 46132555

Email

Events in connection with CARETAKER

• A stakeholder forum is hosted by the IEE conference ICDP 06 "Imaging for Crime Detection and Prevention" on 13-14 June 2006 in London.

<!-- SWrapper // -->

 

eHealth Workshop in Barcelona (Multi-Sector)

The eHealth Seminar is intended to apply the UK experiences implementing cutting edge technology into Health and to enhance the Catalan/Spanish eHealth know how.

Spain - eHealth Workshop in Barcelona

Programme

08:30 Registration

09:15 Welcome to the seminar. Her Britannic Majesty’s Consul General, Mr Andrew Gwatkin

SESSION 1

09:30 NHS “Connecting for Health” Programme

Community Care and Technology

10:30 Company 1: Case study

11:00 Company 2: Case study

11:30 Coffee break

SESSION 2

Interoperability Across Health Domains

12:00 Company 3: Case study

12:30 Company 4: Case study

SESSION 3

Digital Imaging

13:00 Company 5: Case study

13:30 Company 6: Case study

14:00 Networking lunch

SESSION 4

Healthcare Standards/Clinical Governance/Security Information Management/ Patient Safety

15:00 Company 7: Case study

15:30 Company 8: Case study

SESSION 5

16:00 The future: eHealth strategy for 2011-2014. Guru from the UK

16:30 Coffee break

SESSION 6

16:45 Discussion Panel

17:15 One to one meetings

18:00 Closing Seminar & End of Seminar activities

2010 NEM Summit 13-15 Oct. 2010 - Barcelona CATALONIA

As you may already know, the 2010 edition of the NEM Summit is set to be a not-to-be-missed event for all those interested in the Future Internet and in the fast paced evolution of the networked and electronic media industry.

Beyond a plenary session where senior industry leaders and high-level officials from the European Commission will present their views on the NEM research and industry landscapes, the event will feature four parallel tracks covering the technical, economical, and societal aspects of networked media and services:

• Green NEM
• Content Delivery
• User Centric Content Technologies   
• Innovative Media Services

A call for papers has just been launched
http://nem-summit.eu/call-for-papers
and we invite any interested stakeholders to submit scientific, technical or business-oriented papers.

The event will leave time for networking, providing in particular a key opportunity to meet partners and make final adjustments for proposals to be submitted at the first Future Internet PPP Call.

The event will also include a 2000 m2 exhibition area devoted to showcases and demonstrations of research results from key European and global players in the NEM area.
Cross-fertilization will be encouraged, namely in the framework of the first "electronic art contest" that the NEM European Technology Platform has just announced.
And much more for the 500+ awaited delegates!

Time has come to register, consider to exhibit, and submit a paper (submission deadline: July 21, 2010). EU projects are also offered the opportunity to answer a call for exhibition.

We would be pleased to answer any queries you may have or provide you with any further information you may need.
Looking forward to welcoming you in October in Barcelona.

Florent Genoux,
On behalf of the NEM Summit Organising Committee

The NEM Summit is organised by the NEM Initiative under the aegis of the European Commission with the support of the Spanish eNEM platform
Platinum Sponsors: BBC, Technicolor
Gold Sponsors: 3DLife, BCE, Tecnalia, Orange, Telecom Italia, STMicroelectronics, Intel, INRIA, IRT, NextMedia

FI-PPP European Information Day. BRUSSELS, 8 July 2010, 9:30-17:00

Hotel Crowne Plaza Brussels - Le Palace
Rue Gineste 3
B-1210 Brussels, Belgium
T: +32 (0)2 203 62 00
F: +32 (0)2 203 55 55

Registrations and contributions 

The event aims at giving a detailed briefing to interested stakeholders about the FP7 ICT work programme 2011/12 objective 1.7 to 1.10 which related to set-up of the public-private-partnership (PPP) on the Future Internet. In addition, the workshop will be a platform for potential proposers to exchange their ideas and to network.
We assume that participants to this event will have read the FP7 ICT work programme text, in particular all objectives related.
This information day on the FI-PPP will be followed next day (i.e. 9 July 2010) by an information for the Public Private Partnerships' Energy-efficient Buildings, Factories of the Future and Green Cars. Please note that you will have to register separately for both events.

Future Internet Public Private Partnership (FI-PPP)
The growing complexity of the Internet fuelled by the current and future blossoming of services and usages,the connection of trillions of "things",the exacerbation of security and privacy threats,the advent of a "mobile Internet",and the trend towards personalisation of content and services,calls for a decisive action aiming at enhancing or even reshaping the Internet.
Against this background,the Future Internet Public Private Partnership (FI-PPP)is Europe's concerted action to enable a new wave of Internet-enabled industrial spin offs and to ensure that the future evolutions of the Internet will meet the expectations of an increasingly connected society.It puts forward consistent phases,under a single,structured and sustained initiative,in order to provide European Internet stakeholders with the opportunity to investigate different technological options and to put them at work to address the irreversible policy "megatrends"to make economies,business processes,infrastructures and societies "smarter"and more sustainable.
The objectives under this Work Programme denote the first two phases of this PPP under
Framework Programme 7 with two dedicated Calls,which are envisaged to be continued by a
third phase and Call under Work Programme 2013.
The goal of the FI PPP is twofold:a)to strengthen and further expand the competitive
position of the European ICT industry,particularly telecommunication operators,mobile
devices manufacturers,software and service industries,content providers and media;b)to
contribute to the irreversible policy trend towards a more sustainable society,by
demonstrating that key business processes can be made smarter through tighter integration
with Internet capabilities.In order to help closing the traditional gap existing in Europe
between R&D and innovation,the FI PPP combines a medium-term "market pull"approach
driven by the needs of applications,with a "technology push"approach matching the research
agenda of major European technology suppliers.The resulting ecosystem is expected to bring
together the demand and the supply sides,and will allow involving users early into the
research lifecycle,thus contributing to shorten time to market of products and services.There
is a high potential for exploiting synergies between the use-case related activities under this
PPP and the projects under application-oriented Challenges and Objectives of the ICT
Programme.
The FI PPP will pave the way towards a new Internet that will address the technical shortcomings of today's infrastructure,while promoting innovation and competition;allow Europe to effectively respond to the emerging societal challenges be it in terms of transport services,green energy and sustainable development or health services;and provide economicഊactors with opportunities to devise and launch new business ventures in the above application
sectors.
The implementation of the FI PPP is targeted to enable early availability of a framework
(specification,standards,implementation and research/usage validation trials)to ease and
speed up the development,deployment and take-up of novel and innovative added-value
services in Europe,based on Internet-enabled smart infrastructures.The main result of the
Initiative will be a pan European demonstration of Future Internet services.This will target
facilitation of an early uptake of the results in the European marketplace,benefiting European
citizens,and a competitive advantage for European industry in the global market place.
Typical targeted validation environments include (without being restrictive to)“Integrated
Smart Cities ” or "smart regions"as catalyser and federating theme for the FI PPP programme
and projects,and also as focal points of the public contribution to the initiative.Future “Smart
Cities ” or regions will be penetrated by a multitude of Internet technologies enabling real time
communication and processing of massive volumes of data making it possible to cover
innovative applications and typically covering several use cases.
The PPP will pursue an industry-driven,holistic approach encompassing R&D on network
and communication infrastructures,software,service and content/media technologies and
their deployment on real application contexts.The main technical outcome of the PPP will be
a versatile (multi-use case)and open communications and service platform,complemented by
reusable,generic and commonly shared technology enablers.The platform will be used by
many actors,in particular by SMEs and public administration services,to develop new and
more competitive services,to enable more flexible business processes and more sustainable
public infrastructures and utilities.Contribution to EU policies is expected to be through
focused pilot applications having a public local/regional dimension,e.g.in smart cities or
regions.Strong efforts are expected related to openness,standardisation,and certification.
The objective is composed of four tightly related Objectives.The overall budget foreseen
under this Work Programme is EUR 170 million with the budgets per Objective being only
indicative.Projects to be selected under the Objectives outlined below are expected to
collaborate under a common governance structure at programme level (see Annex TBC).
Object ve FI.ICT-2011.1.7 Technology foundat on:Future Internet Core Platform
Target outcomes
Design,development and implementation of a generic,trusted and open network and service
platform making use of and integrating advanced Internet features enabling take up in
innovative "smart applications".This includes the specification of open interfaces from this
Core Platform to domain-specific instantiations addressed by projects under the "Use case
scenarios and pilots"activity.The target platform may draw upon resources from several
independently controlled domains through ad-hoc aggregation of resources.The target
platform can be generically reused for multiple user contexts and to develop the "smart
applications"corresponding to specific use cases,with an important focus on use cases of
public/social or economic value.In the second and third phase of the PPP the implementation
of prototypes of the core platform across the testing infrastructures made available by the
"Use case scenarios and pilots"activity will form the baseline for large-scale experimentation
and validation across multiple use case scenarios with widely distributed geographical
coverage.ഊThe major innovation in this research and development work is expected from the engineering
and scaling-up of advanced Internet technologies currently researched in Europe,enriched by
the necessary integration and functional components,enabling a comprehensive capability for
generic and domain specific services and applications.The work should take a comprehensive
system view of the Internet,underpinning the convergence of network-centric approaches of
operators and telecom equipment manufacturers with web-based and service-oriented
approaches of the software and service providers and integrators.Generic Enablers are a key
feature of the Core Platform.They offer functionalities that can be reused and composed for a
multiplicity of use cases and are at the heart of the platform versatility.Core platform
functionalities should include:
-the general capability to draw upon resources from several independently controlled
domains through ad-hoc aggregation of resources;
-upgraded network capabilities covering requirements derived from innovative Internet use
cases and from the operational needs of smart infrastructures;
-information processing capabilities enabling the generation,composition sharing and
exploitation of huge amounts of data in support of context aware applications and
enabling "mash up"applications (e.g.running on top of different use case-specific
instantiations);
-generic service infrastructure capabilities enabling application-related services,"things"
and contents to be visible and accessible by end-users within and across domain-specific
instantiations in a uniform way enabling "services mash-up";
-real time application capabilities based on coupling sensor and actuator networks to the
internet,through a uniform reference architecture;
-adapted network/service management schemes including traffic flow optimisation,trust
and security;
-trust and identity capabilities enabling end-users,devices,digital objects and service
providers to be identified globally and across multiple domains in a trusted manner;
-use case-independent application and service development tools including application
programming interfaces and software development kits;
-Where and if appropriate,platform federation and interoperability between platforms or
instantiations thereof from an architectural perspective and beyond data integration.
The dynamic specification of the Core Platform functionalities largely depends on the
requirements stemming from the identified use cases.In addition,the testing infrastructure,on
which the core platform is to be implemented will be provided by the "Use Case scenarios and
pilots"activity.Therefore,an efficient collaboration with the projects generated under the
"Use Case scenarios and pilots"(see b)below)is a mandatory requirement.It is in particular
expected that the chief architect of the core platform chairs an architectural board with
participants from all other activities.
The Target Outcome of this phase covering phases 1 and 2 of the PPP includes:
i)System design:through research covering the specification and design of the
functionality and interfaces of the core network and service platform;ഊii)Early prototyping:the phased development and maturing of a reference
implementation with a convincing subset of the targeted capabilities of the Future
Internet Core Platform;
iii)Early implementation and validation:the provisioning of the Core Platform on a
medium scale pan-European FI testbed infrastructure supporting use case specific
experiments.
Funding schemes
One IP
Indicative budget distribution and duration
-EUR 40 million;a minimum of 30%of the budget is expected to be kept flexible for
distribution among partners complemented by Open Calls to allow for responding to
emerging user needs not known from the outset.
-Duration:3 years
Call
-FI-PPP 2010

Another call under this l ne of act on is planned under WP 2013,which s expected to address
the extens on and matur ng of the Core Platform prototypes and the r mplementat on on the
large scale FI testbed nfrastructure.At this stage,the final outc me of this l ne of act v ty
will be a reference mplementat on of all targeted capabil t es n an operat onal prototype of
the Future Internet Core Platform,on which appl cat on domains have built the r domain-
spec f c instant at ons and run large scale demonstrat ons.

Objective FI.ICT-2011.1.8 Use Case scenarios and pilots
Target outcomes
The work focuses on vertical use case scenarios whose efficiency,sustainability and performances can be radically enhanced through a tighter integration with Internet based advanced network and service capabilities.The target use cases should cover innovative applications scenarios with high social or economic impact making used of advanced Future Internet capabilities.Without being restrictive,examples of such target use cases expected to be elaborated in dedicated projects include systems

-for utilities like the electricity grid,
-for communication,
-for traffic and mobility management,
-for healthcare,
-and for ubiquitous access to networked digital media.

Each proposed use case is expected to make use of technologies and functionalities
leapfrogging current innovative internet technologies,such as
-context awareness and sensor networks,ഊ-advanced real time information processing capabilities handling huge volume of information,
-ad-hoc service composition and mash ups,
-managed broadband connectivity and services,
-embedded media support for interfaces easing the interpretation of processed contextual data.

The work covers a full R&D&I cycle,including use case characterization;specification of
platform requirements;development;technological validation prototypes,and large scale pilot
demonstration/validation.Of particular importance for each selected use case is the
identification of usage specific requirements versus generic requirements that can be
implemented through common technological enablers.The latter will be developed by the
"Core platform"activity which takes a central role in collecting requirements and defining
generic enabling capabilities and interfaces,feeding them back into the specifications for the
usage area pilots.The definition and preparation of the pilot sites shall be based on the
provisions made by the "Capacity Building and Infrastructure Support"activity.A pan
European approach is targeted for the implementation of pilots and validation trials.
These sequential activities will be split over two phases:

Target outcomes Phase one:
i)A comprehensive set of detailed technical,functional and non-functional
specifications for a pilot in the given usage area:
-use case characterisation;
-architectural model outlining how the infrastructure supporting the use case
interworks and integrates with the PPP core platform and the Internet in general;
-framework of functional and non-functional specifications,requirements,
technology enablers and use cases;
-design of the domain-specific instantiation of the core platform consisting of a
selection of functionalities complemented by domain-specific capabilities,
including the definition of open interfaces and interoperability requirements.
ii)Development of domain-specific capabilities and conceptual prototypes
demonstrating critical technological solutions and the overall feasibility and approach
suggested for Phase 2.
iii)A Phase 2 implementation plan,which also includes a detailed analysis of the
pilot sites and the locally provided infrastructures,and a plan for consortium and user
community building.
Target outcomes Phase two:
i)Working pilots and test-beds building upon common components and enablers as
provided under the Core Platform activity and covering the selected pilot sites.
ii)Selected test applications implemented on pilot sites.ഊiii)Validation of the openness and versatility of the core platform and its software
development kid,through implementation of mixed use case scenarios originating
from more than one use case project.
iv)A detailed plan for how to move into Phase 3 in which a massive expansion of the
platform usage is envisaged,facilitated by local and regional stakeholders including
SMEs.
i)Phase One
Funding schemes
-up to 8 IPs
Indicative budget distribution and duration
-EUR 5 million per use case project
-Duration:max 24 months
Call
-FI-PPP 2010
ii)Phase Two
Funding schemes
-up to 5 IPs,i.e.consolidation of up to 5 different use cases
Indicative budget distribution and duration
-13M€ per use case project;a minimum of 10%of the budget is expected to be allocated
through Open Calls to allow for local solution providers and system integrators to get
involved.
-Duration:max 24 months
Call
-FI-PPP 2011

Objective FI.ICT-2011.1.9 Capacity Building and Infrastructure Support
Target outcomes
The goal is to leverage existing public investments in advanced infrastructures to support the
deployment of advanced pilots demonstrating the versatility of the Core Platform across a
multiplicity of heterogeneous environments.Several European regions or cities are
increasingly becoming equipped with advanced infrastructures (e.g.sensor platform,
broadband islands..).The objective of this activity is hence to identify,taking a pan European
perspective,those infrastructures that could eventually be integrated with the platform to
support the large scale trial phase,and to identify the related interoperability requirements.
These interoperability requirements will also help the definition of 'common enablers',under
the "core platform"activity,as they will drive the required level of virtualisation making it
possible to seamlessly integrate various heterogeneous infrastructures and to federate them
according to use case requirements.This activity requires putting in place a partnershipഊstrategy with the infrastructure owners and a detailed understanding on the operational usage
taking into account that these supporting infrastructures will be used in "shared"modes.
Finally,supporting infrastructures need to be upgraded according to research results driving
additional requirements and constraints to support the target use cases
The Target Outcome of this activity consequently covers:
1.The identification of existing and future advanced test and experimental infrastructures
across Europe and the associated technological constraints that need to be overcome to use
these for conducting large scale (ultimately user driven)trials of new innovative and
integrated Future Internet Applications.As typical – but not exhaustive -examples these
infrastructures may include sensor platforms,advanced broadband wireless networks,
server farms and service environments,energy grids,content delivery networks
2.The maintenance of a web based repository of available infrastructure potentially engaged
in trials within this initiative and of their key functional characteristics;
3.The identification of the usage-related operational constraints derived from these
infrastructures;
4.The interoperability requirements fuelling the "common enabler"definition of the core
platform and making it possible to assemble a pan-European federation of test and
experimental infrastructures where the results of the Initiative can be implemented.
5.The controllability of key functionalities of these infrastructures across Europe through
the common APIs/SDK of the core platform by the start of phase 3 so that the application
providers from the industry sectors can validate their application scenarios in a
representative environment.
6.The necessary adaptation and validation of these infrastructures in view of supporting
additional usage requirements stemming from the selected pilot use cases and a mix of
those.
7.The integration of these infrastructures within the selected pilot use cases targeted in
phase 2 (it is expected that targeted use cases are implemented through specific islands
only requiring integration of a subset of all available infrastructures).
Funding schemes,indicative budget distribution,Calls:
i)to cover items 1.to 3.above:
Funding schemes
-One CSA
Indicative budget distribution and duration
-EUR 3 million
-Indicative duration:3 years
Call
-FI-PPP 2010 (starting from phase 1)
ii)to cover items 4.to 7.above:ഊFunding schemes
-One IP
Indicative budget distribution and duration
-EUR 12 million
-Indicative duration:2 years
Call
-FI-PPP 2011 (starting from phase 2)

Objective FI.ICT-2011.1.10 Programme Management and Support
Target outcomes
The implementation of the PPP activities across a limited set of interrelated projects requires
the setting up of a comprehensive management and support organisation.Beyond pure
management and co-ordination issues,the objective of this activity is to address the non
research activities that are needed for a successful implementation of the PPP.
The objective is also to prepare for the large participation of SME's at the level of the
validation &large scale trial phases.As the pilot experiments will be implemented following
a "user driven innovation"scheme,SME's will be targeted in priority to develop the
validation experiments running across the target platform and for the selected use cases.
Participation of SME's to this phase hence needs to be planned sufficiently in advance.
The Target Outcome of this activity covers:
· The co-ordination and planning of the necessary flow of information across the various
projects implementing the PPP;the management of dependencies and synchronisation of
the project activities across the programme;
· Planning of trials and large scale experiments;platform operator function (scheduling and
planning)
· Support and coordination of the necessary standardisation stemming from the openness
requirements of the target platform;set-up a certification programme for relevant enablers
and services;coordination of PPP contributions to international standardisation as to
maximise impact in terms of industry in Europe capitalising on the results;
· Develop,agree with participating project,implement and operate the necessary and
adequate programme governance structures for an efficient and effective implementation
of the programme,catering for effective day-to-day management of programme execution
and for advisory roles.To this regard,programme governance models developed by joint
undertakings such as FCH,CLEANSKY or ARTEMIS might be used as examples to
build on.It is expected that all programme activities adhere to the programme governance
structures;
· Contributions related to regulation and EU policies made to the relevant bodies;support to
the necessary regulatory evolution making it possible to operate such a distributed
platform across Europe,with a perspective of an internal market for trusted and secure e-ഊservices for data repositories,in particular for what concerns use cases related to public
sector priorities;
· Identification of SME's participating in the trials of phase 3,training programme and
support to application developments.
· Co-ordination of dissemination and awareness activities,including dissemination
activities towards European cities and regions,awareness raising actions targeted at policy
makers responsible for local or regional developments;co-ordination of participation to
large events like international exhibitions and fairs;preparation of high quality
dissemination material;etc.
· Set-up and co-ordination of programme boards in agreement with participating projects.
· Maintenance of a programme-level IPR regime.
Funding schemes
-One CSA
Indicative budget distribution and duration
-EUR 10 million
-Indicative duration:3 years
Call
-FI-PPP 2010

Expected Impact of the FI PPP (The 4 objectives described above)
· Significant increase of the effectiveness of business processes and novel approaches to
the operation of infrastructures and applications of high societal value.This will be
supported by a reappraised internet architectures,services and technologies in large-
scale application contexts;
· Reinforced industrial capability on novel service architectures and platforms,building on the longer-term requirements of the internet and encouraging European players to embrace the challenges of smart infrastructures,whilst contributing to EU policies in terms of innovation,sustainable growth,energy and environmental targets;
· New opportunities for novel business models based on cross-sector industrial partnerships built around Future Internet value chains,involving users and public authorities at local,regional and national levels,and providing SME players with
opportunities to offer new products,equipments,services and applications.
· Creation of new European-scale markets,overcoming potential fragmentation,for smart infrastructures,with integrated communications functionality,contributing to economic growth and to European leadership in global ICT applications markets.
· Evolution (not clean slate)of Future Internet infrastructure compatible with the emergence of open,secure and trusted service platform for building networked applications that can be leveraged through user-centred open innovation schemes;ഊ· A comprehensive approach towards regulatory and policy issues such as interoperability,openness,standards,data security and privacy within the context of the Future Internet complex and ‘smart ’ usage scenarios.This may also address the
required methodologies,procedures and best practice needed to address transnational aspects where a high degree of public-private co-operation is needed.Participation of the public sector in the PPP will be a key asset to progress in these non-technological issues.

The virtual conference features event processing luminaries, early adopters and experts:

• W. Roy Schulte, Vice President and Distinguished Analyst, Gartner
• Opher Etzion, IBM Senior Technical Staff Member and chair of the Event Processing Technical Society
• Christopher Bird, Chief Architect at Sabre Airline Solutions
• Paul Vincent, CTO Business Rules and CEP, TIBCO Software
• Colin Clark, Chief Technology Officer, Cloud Event Processing, Inc.

Learn how Event Processing enables agencies and corporations to profit from continuous intelligence.

Hear from industry pioneers, leading vendors, and early adopters on Event Processing technologies and techniques that increase mission and business visibility and responsiveness.

Interact with industry experts, leading adopters, and peers via question and answer segments, and follow-on community discussion.

Influence, participate in, and benefit from the rise of event processing as we launch the Event Processing Community.

Featured Sessions

Smart Systems and Sense-and-Respond Behavior: The Time for Event Processing is Now
W. Roy Schulte, Vice President and Distinguished Analyst, Gartner
 
The need for situation awareness and sense-and-respond behavior is no longer limited to niche applications. Virtually every large new system in business, defense, and government is becoming event-driven in some aspects of its operation. This overview session will introduce the fundamentals of event processing, and explain where it is used and why it is a fundamental part of smart devices and smart applications of all kinds.

• How does continuous monitoring enable situation awareness and provide earlier detection of threats and opportunities?
• What design patterns form the basis of smart devices and event-driven processes?
• Which technologies and product categories are applied to different event processing usage scenarios?

Analyze, Sense, and Respond: Identifying Threats & Opportunities in Social Networks
Colin Clark, Chief Technology Officer, Cloud Event Processing, Inc.

Colin Clark will demonstrate the use of streaming map/reduce and complex event processing to identify threats and opportunities using Twitter. This session will identify specific technologies, how they're inter-related, and how the DarkStar and Telescope products from Cloud Event Processing can be used to do this quickly and efficiently utilizing elastic resources.

• What are short lived trends in Twitter and how can they be taken advantage of?
•  What are people saying about your company on Twitter and is it positive or negative?
• How do Twitter usage patterns differ geographically and what opportunities does this present?

Case Study: Event Distribution Architecture at Sabre Airline Solutions
Christopher Bird, Chief Architect at Sabre Airline Solutions

Christopher Bird will present the event distribution architecture in use within Sabre Airline Solutions. This event distribution architecture is used as the backbone for delivering events and content through the high volume, reliably network within Sabre.

Of special interest is overcoming the following issues:

• Large payload delivery
• Validation in the delivery network
• Situational awareness
• Management of control structures
• Message redelivery and failure semantics
• Use of the event network for test data distribution

Event Processing - Seven Years from Now
Opher Etzion, IBM Senior Technical Staff Member and chair of the Event Processing Technical Society

While event processing is considered as an emerging technology in enterprise computing, it has barely scratched the surface of its potential. This talk will illustrate a world in which event processing is everywhere, consumed by everybody, and used for enterprises as well as consumers.

First, we survey six trends:

• Going from narrow to wide coverage
• Going from monolithic to diversified architectures
• Going from propriety to standards in various areas
• Going from programmer-dependent application development to user independence in application development
• Going from event processing as stand-alone technology to event processing as pervasively embeddable technology
• Going from using event processing reactively to using event processing also proactively

For each of these trends a roadmap will be described as well as the target goals. In addition there will be a discussion on four areas which require development beyond the current state of the art in order to realize the "event processing anywhere" vision: event processing virtualization, event processing software engineering, event processing scalability and intelligent event processing, in each of these topics we discuss the challenges and directions, and the way they play in the bigger picture.

Events, Rules and Processes - Exploiting CEP for "the 2-second Advantage"
Paul Vincent, Chief Technology Officer, Business Rules and CEP, TIBCO Software

Complex Event Processing technologies are normally associated with higher performance investment banking decisions - providing a trader with an edge over the competition. However, the advantages of the simplicity and performance of the "event processing" ideal has also attracted the attention of other businesses requiring processes and services with very high throughput yet handling multiple scenarios and policies.

Many of the ideas present in the BPM and SOA state of the art are also present in CEP systems: event bus, rules engine for decisions, model-driven development and such. Additionally, CEP adds a focus on "event", "time", and their associated pattern-based processes - which possibly better reflects the cognitive processes we undertake ourselves.

In this session, we explore some of the design patterns and case studies that exploit CEP technologies to provide adaptive event-driven processes and services. As well, we discuss why the architects deviated from the safe road of "the orchestration diagram and BPEL" to the more rare declarative rules and the other models associated with event pattern detection - for SOME of their processes and services!

Hosted by

Sponsorships Available

<!--webbot bot="HitCounter" i-image="5" I-ResetValue="0" I-Digits="0" U-Custom="custom.gif" startspan -->

SoundICTs: Endowing human auditory capacities and embodying audition faculties into artificial systems

Ferran Cabrer I Vilagut - consen @ consen.org - MUFICATA s.l. (CONSEN EEIG - ACAIA.org)

BRIEF EXPLANATION
SoundITCs
Proof-of-concept ICT models for endowing human auditory capacities and embodying audition faculties into Artificial Intelligent and robotics systems for multi-sensory cognition and reasoning and natural personality, adaptation and evolution
Endowing and embodying Human Audition faculties into robotics systems investigates, develops and proof-of-concept ICT models of acoustic, cognition and robotic systems to leverage the rich though untapped potential of the information associated to sounds, by simulation of human audition perception and understanding and multi-sensory reasoning and population of evolutionary capacities, to embody consciousness faculties, empathic personality and natural evolution into robots.

AMBITION
ACAIA promising visionary idea is to exploit the information associate to sounds for all Artificial Intelligent system and the benefit of the Knowledge Society.
ACAIA Identifies a unifying scientific goal, Audition for all Robots
ACAIA problem is that the justification of the installation of audio-sensors networks for acquiring sounds is the auditory capacity, or the power to extract the useful associated information and the ability to understand it. The complete audition faculties is reached with the multimodal reasoning that embodies consciousness to the robot and facilitates building natural and human robots.
ACAIA solution is distributing and sharing development and experience on Internet to be available for all robots from semantic repositories of Acoustic Processing, Multi-sensory Cognition and Robotics Reasoning, forming and populating two meshes of Artificial Sensing, Thinking and Acting and systems learning, teaching, training and factoring.
ACAIA exploring and modelling concepts and technological of Acoustics processing, Artificial Audition, Cognition and Robotics that can contribute to Multi-Modal sensing and reasoning and Consciousness of robotics challenges of long term importance for Europe.
ACAIA topics are the following:
Understanding The Human Audition, Multimodal Reasoning and Consciousness through proof-of-concept of ICT model.
Personality: Conviction, Anticipation and Intuition by simulation Acoustic Processing, Audio, Cognition and Robotics Technologies Maturing emerging ACAIA community for future problem solving and leading Technologies Robotics for Society
-------------------
The human audition for robotics project will have the following objectives:
- to extract all the genuinely meaningful of sounds and noises;
Contributions Flagship Consultation 2010 14
- to generate accurate “quasi real-time“ audio-monitoring and audition understanding and continuous learning of “robotics systems”;
- to deal audio-raw-data, as “audio streams as film” formed by a sequence of frames of 120ms approx and the mathematical algorithms for - filtering - detection of Acoustic Events, as graphic pictures (frequency x intensity). The audio multi-stream processing device “hears” is filtering 6-8 frames per second for each stream (cinema 24fps, TV 50fps and some video-games 72fps);
- to implement two audio sensing behaviours: active listening to and passive hearing;
- to facilitate a public space “The Future of Internet of Sounds Services” ecosystem of acoustic resources, semantic repositories and open learning and decision-making algorithms in pervasive environments and structure these libraries as a software development to develop, populate, integrate, package and download applications available for all to supply acoustic retraining and cognition and robotic training. This populated, shared and distributed capabilities form an evolutionary ecosystem have to allow the evolution of robotic systems;

IMPACT
The world of sound is not exploited in the Information Society. Being in the XXI century, and given the importance that humans assign to sound in their daily environment, it is interesting to note how few ICT systems are actually exploiting acoustic information naturally associated with sound and noises, such as time, position, event, source, material, size, composition, risk, environment, velocity, traffic, weather, health, disease, context, activity, occupancy, etc.
The project wants to use all the information coming from the environment to enable robots, which can hear and listen, and which can understand the meaning of the acoustically conveyed environmental information and which can hence provide useful services to the knowledge society in an improved, bio-inspired, way.

INTEGRATION
The SoundITCs scheme stimulates non-conventional targeted exploratory research cutting across Acoustics, Cognition and Robotics disciplines and acts as a harbour for exploring and nurturing new research trends and helping them mature in emerging research communities, such as Acoustic Computing for Artificial Intelligence Applications (ACAIA.org).
Harmonic and close interaction between perception and actuation of the audition and cognition for robotics self-configuring sub-systems that consists of five different subsystems:
1.Acoustic Sensing. Acquiring and processing. Listen to and hear the environment.
2.Acoustic Reasoning. Embodying meaning and understanding audio information.
3.Consciousness: Intuition, improvisation and conviction. Multi-modal perception and reflection. of novel experiences consisting of multi-modal perception.
4.Learning and training. Autonomous active learning and natural “pet” training.
5.Human, natural and empathic behaviour. Internal self-organized top-down control.
The Audition perception of robotic systems become aware of the physical world for modelling and control purposes through the acoustic acquiring of audio-sensor and meaning of audio data and multi-modal reasoning to represent, define and qualify the reality to interpret in the light of a pool of shared-experience on Internet and models learnt of behaviour to decide the actuation.
In particular, the main objectives of the SOUNDICTS project are the following:
1. To develop a conceptual and theoretical framework for human audition system for robotics.
2. To design and implement a computational realisation, processing, classification, analysing and packaging Acoustic Events. Main elements of this development include a meta-structure of resources named semantic repositories, novel graph-processing algorithms facilitating multimodal reasoning and development, standards methods that extract sounds meaning and populate these on Internet based on heterogeneous and distributed sources, and user interfaces that facilitate effective and efficient exploration and use of such repositories.
Contributions Flagship Consultation 2010 15
3. To demonstrate the validity of the framework and to evaluate and assess its properties on the basis of a diverse set of demonstration scenarios drawn from different domains, and a range of hypothetical test scenarios with known characteristics, that will be proposed in the project.
4. To assess whether the methodology and its proof-of-concept implementation could lead to a new modular audition sub-system paradigm.
5. To widely disseminate the results of the project, raise awareness in the relevant communities and explore ways of exploiting the project results.
The main disciplines involved:
Acoustics: DSP, Acoustic processing and Detection of Acoustic Event Mathematics: Mathematical algorithms of DSP, cognition risk management, decision making and learning. ICT Audio-sensors and audio-sensors networks. Semantic and ontology. Artificial Intelligence. Systems Architecture. SOA and GPL Psychology: Cognition. Perception. Multi-sensory reasoning. Personality and behaviour. Learning.Robotics: AmI, modularity, integration, machine-learningOther: statistic, ecology, evolutionary theory, sustainability and other sciences.

PLAUSABILITY
Human audition and cognition for robotics Project structureTo reach the goal, advances in understanding and technology development are needed in four closely interrelated areas:
1.the catalogue of technologies for endowing human auditory capacities to robots;
2.the integration of sub-systems for embodying audition faculties into robotics systems;
3.the modelling of natural cognition and reasoning process for design robotic systems;
4.the multi-senses reasoning to strength the robot consciousness and personality.
These four issues are being addressed in five closely intertwined work packages, which form the core of the project. There are two additional work packages, one devoted to project management WP0 and another to dissemination WP5, both of which will be coordinated by the project manager
The main works are to proof the following simple ICT models of modular subsystems:
WP1 ACQUISITION AND PERCEPTION.
Acoustic acquiring and processing by a simple audio-sensors network (ASN) supplying audio-raw data. This simplification excludes all works in sensors, MEMS and microphones, in interactive process, beamforming, sounds tracking and acquiring sounds from remote environments. The acoustic processing device that receive and process the audio information, via multi-streaming, capable of identify and detect acoustic events for extracting meaning and knowledge and manage “abnormal sounds”;
WP2 RECOGNITION, REFLECTION. AND DECISION.
Acoustic reasoning central unit providing meaning and understanding to the auditory and *-perception, for multi-modal reflection, *-senses cues, risk-management consideration and decision-making of actuations. Algorithms and software systems allowing to parametrise the levels of personal behaviour in the reactions (interaction or actuation) and the adaptation to environment and context for generating autonomy, conviction, intuition and pseudo-natural personality to robots;
WP3 ADAPTATION.
Acoustic self-learning and user-training methods and techniques to enhance the features of robotic subsystems and the complete robot system. WP5 designs and develops truly adaptive autonomous systems capable automatically to learn from the experience of all the rest of robots and to teach the rest of robots – shared on the common IoSS ecosystem. The self-learning is based on the management, process and retraining of “unknown sounds”. The user-training will be focused on the behaviours of WP4 personal behaviour, context and missions;
WP4 EVOLUTION.
Acoustic factoring and evolving Getting inspiration from natural ecosystems, the objective of this work is the development of a highly-innovative theoretical and practical Internet of Sounds Services IoSS framework for the decentralized development, deployment and execution of services for future robotic
Contributions Flagship Consultation 2010 16
systems and sensing networks. The framework will be grounded on a foundational re-thinking of current Acoustic, Cognition and Robotic (ACR) service models and of associated infrastructures and algorithms. IoSS develops and supplies externally new audition capacities - acoustic events, cognition processing and robotics applications - to upgrade the robotic systems and subsystems. The building of IoSS provide the foundation for sustaining the development of auditory capacities and cognitive systems for robotics. The continuous process of individual retraining will generate a natural evolution of individual robots, depending mainly on the personality, the user-training and the living environment;
COMMENTS
SMES:
Muficata s.l. ESP
AKG Acoustics GmbH AUT
RTD:
Fraunhofer-Gesellschaft Forderung der angewandten Forschung FHG-IDMT GER
Italian National Research Council ITA
ACADEMIA:
AG ESIGETEL FRA
University of Uppsala SWE
Technical University of Munich GER
University of Portsmouth GBR
PLATFORMS:
Acoustic Computing for AI/AmI Applications (ACAIA.org) community
In addition to this consortium supported by this experts community the research works in this area is unknown.
Persons:
Ferran Cabrer i Vilagut is the President and Manager of CONSEN Euro-Group and CEO of MUFICATA S.L. He is Industrial and Agricultural Engineer of profession and Master in Business Administration. He comes from different sectors, remarking the management of the department of Research and development of phito-pharmaceutical products into LAINCO. In a personal entrepreneur project funded MUFICATA as ICT Enterprises consulting and Corporate Finance firm to provide Enterprises Information and Technological services. The Internet was the instrument used to start International services and converted in EU information and project management services. After a pair of year he funded CONSEN as Open Euro-Cluster in Information Society Technologies formed by eight SMES operating at European Level. It is one that has participated actively from 2002 in european ICT projects. Actually he is contributing in diverse european projects and collaborate as Vice-Chairman of the Knowledge@work community, EURO_COOP_ICT_DEV Romanian research project for a sustainable cooperation throughout Europe, Member of the European Association of Research Managers and Administrators (EARMA), member of eBusiness Support Network (eBSN), Living-Labs and several more European initiatives and networks in IST as eMobility, ISI, NESSI and NEM technological platforms. From 2005 is promoting ACAIA.org initiative.
Dipl. Ing. Rene Rodigast studied Electrical Engineering in Hermsdorf and in Jena from 1989 to 1994. He worked as a technician in the field of audio and video technology for theater, broadcast, and live events. From 1995 to 2002 he was technical director for professional audio and video applications. Since June 2002 Rodigast has worked at the Fraunhofer Institute for Digital Media Technology in Ilmenau/Germany. He is manager in the acoustic department of the group “Multimedia Systems”. One of his major projects was the first implementation of the Wave Field Synthesis technology IOSONO in a commercial movie theater. Another outstanding success was the development of a new sound system for the world’s largest open-air Floating Stage “Bregenz Festival”, Austria in 2005. Rene Rodigast is member of the International Planetarium Society. Contributions Flagship Consultation 2010 17
Fernando Ferri received the degrees in Electronic Engineering in 1990 and the PhD in Medical Informatics at the University of Rome “La Sapienza” in 1993. He is senior researcher of the National Research Council of Italy. He has been contract professor from 1993 to 2000 of “Sistemi di Elaborazione” at the University of Macerata. He is the author of more than 160 papers on international journals, books and conferences. His main research areas of interest are: Geographic Information Systems, Data and Knowledge Bases, Human-Machine Interaction, User Modelling, Visual Interaction, Sketch based interfaces, Risk Management and Medical Informatics. Contributions

12:00 12:15 Introductions, main ideas and available material

12:15 12:30 SoundITs SHORT STREP

12:30 12:45 FETOPEN Flagship meeting  9-10June2010

12:45 13:00 Round of questions

1) template of LoI of the FETOPEN
2) two stages the first SHORT STREP is anonymous and the second is a traditional EPSS proposal
Now we have to work on this SHORT STREP to be sent in firsts of June
Look at the draft of the proposal sent was is needed for the SHORT STREP
The final STREP has to be a 3ME 3years 6-8 partners STREP
  We have to manage the first feedbacks received to asks some points of the ShortStrep evaluation and made an appointment to discuss the preparation of the STREP. This is usually a short meeting during the workshop.
 
Maybe in this FET-OPEN we will consider SensingITs to deal multi-sensing models for robotics systems, in particular Acoustic and Odour Events Detection (AED) and (OED) in parallel. We have to work in the fitting of IDMT and  IAST competences and works
http://aass.oru.se/~ali/icra10ws/index.html

Short proposal  ICT FET Open Call FP7-ICT-2009-C
Work programme topics addressed  ICT-2007.9.0 FET Open
Type of funding scheme: STREP Small or medium-scale focused research project (SHORT PROPOSAL)

Information technologies and models for endowing auditory capacities and embodying human audition faculties
into robotic systems

SOUNDITS

Date of preparation: 11Jun10 (28Sep10)

ICRA 2010 Workshop

Networked and Mobile Robot Olfaction in Natural, Dynamic Environments