PS ID: PS-ES-75732

Status: Open

Date of last Modification: 31/01/2012

Date of Publication: 31/01/2012

Call Identifier: ICT Call 9 (FP7-ICT-2011-9)

Objective: ICT-2011.8.2 ICT for access to cultural resources

Funding Schemes: STREP

Evaluation Scheme: One step proposal

Closure Date: 17/4/2012


Proposal Outline:

CARMA acronym stands for Cultural Augmented Reality Mobile Application. It aims to develop a context-aware object-oriented cultural tourism mobile application able to enhance the information handled by users from their smartphones in an apparent, efficient and useful manner. Such technological service is to be built to work indoor, and will be able to provide complete, contextual and updated cultural information.

The starting point of the process would be an object (for example an art work exhibited at a museum) the user is interested in, and so he points it out with his phone’s camera. The information to be shown would be collected online, combining various technologies, such as touristic/cultural ontologies and state-of-the-art intelligent search engines and displayed in a multilingual, accessible, user friendly interface using Augmented Reality techniques.

The constant extension and evolution of graphic, processing and connectivity capacities of smartphones has opened a wide range of applications increasingly attractive and handy for users. One of the trends in this market is the introduction of Augmented Reality, currently limited to outdoor scenarios where its application can lean on GPS and compass positioning techniques which can determine the place (building, monument, landscape) the user is focusing in with his mobile device. The project seeks for a technological breakthrough at indoor scenarios through the incorporation of object recognition.

CARMA intends to become a very useful tool for those looking for extra cultural information while visiting places of their interest. Main technologies to be integrated within the project:

- A powerful object recognition engine.

- A user-friendly, augmented, multilingual interface.

- A search engine and a content management system.

- Data mining tools.

The main function of the application is to become a sort of “universal multimedia-guide”, not limited to provide just audio descriptions, available to be potentially used worldwide in thousands of museums, cities and regions with no need of any specific technical infrastructure or to reach agreements with third parties.

On the technical side, the project envisages the generation of an impressive and useful user interface based on Augmented Reality, making available enhanced cultural information of an artistic object in a very accurate way. Provided the application will be specified to work at indoor scenarios (no GPS or compass support), and that there will be no need of specific agreements with the holders of the cultural goods (as a consequence, there will be no possibility of relying on any prepared infrastructure to assist in the object recognition), then a powerful way to identify these objects becomes mandatory.

CARMA will include a combination of different state of art pattern recognition algorithms to fulfil the requirement. These techniques can result quite intensive in terms of computational efforts, and so the possibility of distributing these efforts under a mixed architecture will be analyzed. In other words, part of the calculations will be performed locally in the smartphone and another part will be carried out remotely by the application server (cloud?).

In terms of the use of AR, CARMA will implement a technology able to create and show well-positioned 2 and 3D objects in indoors. Taking into account the fast evolution of mobile phones, both at HW and SW level, combined with a surge in application storefront releases, deployments of higher-capacity network infrastructure and recent developments in positioning technologies, a new indoor LBS with AR technology will be considered.

From the side of the search engine to collect relevant data from the web, CARMA will implement crawlers that browse the World Wide Web in a methodical and automated manner. This system will be able to identify and retrieve the required information from different sources in real time, utilizing ants, automatic indexers and web spiders. The importance of a page for a crawler can also be expressed as a function of the similarity of a page to a given query. The main problem in nowadays technology that implements crawling is the web context. That’s because it is very difficult to predict the similarity of the text of a given page and the query before actually downloading the page.


Partner profile sought


Required skills and Expertise:

Experts in:

Image and pattern recognition

Detection, estimation and decision theory

Artificial Intelligence development and modelling. Semantic web and ontologies

Data mining

Augmented reality user interfaces

Development for embedded (mobile) systems

Description of work to be carried out by the partner(s) sought:

Profile 1: Participate on tasks related to the accurate identification of the artworks to be described and/or their relative position within the mobile screen

Profile 2: Participate on the implementation of the augmented reality user interface

Profile 3: Participation on the development of all the stuff related to the gathering of relevant information of a given piece of art: semantic web crawlers, data mining, etc

Type of partner(s) sought:

Universities, Technological institutes or Research intensive companies with a proven and remarkable track record in semantic web, ontologies, data mining techniques, natural language analysis and similarity prediction

Universities,Technological institutes or Research intensive companies with a proven and remarkable track record in image/pattern recognition and other innovative identification methods

Looking for a Coordinator for your proposal: Yes


Si está interesado puede contactar con: