Skip to main content
Aller à la page d’accueil de la Commission européenne (s’ouvre dans une nouvelle fenêtre)
français fr
CORDIS - Résultats de la recherche de l’UE
CORDIS
Contenu archivé le 2024-05-27

Artificial vision systems based on early cognitive cortical processing

Objectif

Visual scene analysis in higher vertebrates reaches a level of performance, in terms of accuracy and efficiency, unsurpassed by man-made artefacts. The goal of the ECOVISION project is to use knowledge of biological systems to build a hybrid hardware-software visual scene analysis system with early-cognitive properties. Early-cognitive systems utilize implicit knowledge of the world. The main innovative aspect of this research is the creation of a distributed processing architecture that emulates self-adaptive early cognitive analysis at the level of psychological "Gestalts" (systems which embody statistical regularities of the world) utilizing dynamic recurrent feedback and predictive mechanisms to generate a self-emergent level of analysis. Visual scene analysis in higher vertebrates reaches a level of performance, in terms of accuracy and efficiency, unsurpassed by man-made artefacts. The goal of the ECOVISION project is to use knowledge of biological systems to build a hybrid hardware-software visual scene analysis system with early-cognitive properties. Early-cognitive systems utilize implicit knowledge of the world. The main innovative aspect of this research is the creation of a distributed processing architecture which emulates self-adaptive early cognitive analysis at the level of psychological "Gestalts" (systems which embody statistical regularities of the world) utilizing dynamic recurrent feedback and predictive mechanisms to generate a self-emergent level of analysis.

OBJECTIVES
ECOVISION will build a visual scene analysis system with neuronal architecture guided by cognitive context, which operates in video-real time, thereby providing the first entry point to a new generation of artefacts with a more human-like pattern of intention driven behaviour. ECOVISION:
1)contains dynamic feedback loops leading to self-organization of neuronal activity;
2)embeds early cognitive analysis at the level of context adaptive visual receptive fields ("visual cortical RFs");
3) defines a dynamically changing "map-like" representational structure of these RFs weighted by their momentarily existing task-relevance;
4) implements predictive mechanism to generate a self-organizing level of analysis;
5) front-end pre-processing steps are studied, evaluated and adapted to suit in FPGA technology in order to operate in video real-time.

DESCRIPTION OF WORK
ECOVISION consists of three main project parts and focuses on a self-navigating system (like a car).
Project Parts: A) Design of an FPGA-based front end for the computation of the motion structure in the visual scenes in video real-time. (The consortium possesses already such a device for the extraction of stereo-depth information);
Project Part B) Design of a high level processing software device with neuronal structure, which analyses the front-end information with respect to its intrinsically existing visual context.
"Visual context" is the embedding of Gestalt-laws, established by psychology and also in cortical neuronal processing, in a computer vision system. "Gestalts", thereby, represent pixel groups, which have shared statistical properties in space and time. Thus, they define segments in space and time and can, therefore, be captured by highly-adaptive receptive fields (RFs) of ECOVISION's processing units. *Control is intention driven* in that every scene will be analysed with respect to an intentional context within which the momentarily existing Center of Interest (CoI) is determined by the dynamic structure of the visual scene. To this end, the RFs of the units shall be arranged on a map-structure that has an enhanced resolution at the CoI. This map, thus, represents, the momentarily existing visual relevance and the CoI can be regarded as a flexible mental fovea. To get a deeper insight, these concepts will also be ascertained by biophysically realistic models of visual cortical function;
Project Part C) Design of the complete ECOVISION system by joining the components from A and B.

Champ scientifique (EuroSciVoc)

CORDIS classe les projets avec EuroSciVoc, une taxonomie multilingue des domaines scientifiques, grâce à un processus semi-automatique basé sur des techniques TLN. Voir: Le vocabulaire scientifique européen.

Vous devez vous identifier ou vous inscrire pour utiliser cette fonction

Programme(s)

Programmes de financement pluriannuels qui définissent les priorités de l’UE en matière de recherche et d’innovation.

Thème(s)

Les appels à propositions sont divisés en thèmes. Un thème définit un sujet ou un domaine spécifique dans le cadre duquel les candidats peuvent soumettre des propositions. La description d’un thème comprend sa portée spécifique et l’impact attendu du projet financé.

Appel à propositions

Procédure par laquelle les candidats sont invités à soumettre des propositions de projet en vue de bénéficier d’un financement de l’UE.

Données non disponibles

Régime de financement

Régime de financement (ou «type d’action») à l’intérieur d’un programme présentant des caractéristiques communes. Le régime de financement précise le champ d’application de ce qui est financé, le taux de remboursement, les critères d’évaluation spécifiques pour bénéficier du financement et les formes simplifiées de couverture des coûts, telles que les montants forfaitaires.

CSC - Cost-sharing contracts

Coordinateur

THE UNIVERSITY OF STIRLING
Contribution de l’UE
Aucune donnée
Adresse

FK9 4LA STIRLING
Royaume-Uni

Voir sur la carte

Coût total

Les coûts totaux encourus par l’organisation concernée pour participer au projet, y compris les coûts directs et indirects. Ce montant est un sous-ensemble du budget global du projet.

Aucune donnée

Participants (7)

Mon livret 0 0