Reconstructing the Past: Artificial Intelligence and Robotics Meet Cultural Heritage

Informazioni relative al progetto

RePAIR

ID dell’accordo di sovvenzione: 964854

DOI

10.3030/964854

Data della firma CE 21 Dicembre 2020

Data di avvio 1 Settembre 2021

Data di completamento 31 Ottobre 2025

Finanziato da

• EXCELLENT SCIENCE - Future and Emerging Technologies (FET)

Costo totale

€ 3 522 010,00

Contributo UE

€ 3 522 007,00

Coordinato da

UNIVERSITA CA' FOSCARI VENEZIA
Italy

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

Other (3)

Integration guidelines and plan 3

Describes the 3rd version of the guidelines and plans needed for a successful integration of the different components of the RePAIR system.

Integration guidelines and plan 1

Describes the 1st version of the guidelines and plans needed for a successful integration of the different components of the RePAIR system.

Integration guidelines and plan 2

Describes the 2nd version of the guidelines and plans needed for a successful integration of the different components of the RePAIR system.

Documents, reports (13)

Setup of integration and sharing repository

Describes the setup of the repositories to be created to facilitate interaction during the development phase

Fragment recognition and pose estimation

Describes the work done in WP2.2 where we shall develop matching algorithms to recognise and estimate the pose of the fragments, given possibly partial views and different sensor modalities.

Bi-manual motion planning

Describes the work done in WP5.2 whew we will develop efficient motion planning techniques for the bi-manual robotic system.

Acquisition and 3D modelling of fragments

Describes the work done in WP21 which will focus on deriving an accurate 3D model for each archaeological fragment together with a mapped texture so as to obtain appearance and geometrybased descriptors that will be used by the puzzlesolving and grasping modules

Hand pose selection for grasp planning using vision sensors

Describes the work done in WP51 which aims to learn by explorationimitation the parameters of the visual servoing controllers for softhands

Adaptable bi-manual manipulation platform

Describes the work done in WP4.2 which focuses on the development and control of a dedicated twin robot arm to execute the manipulation tasks of the fresco pieces.

Quality Management plan

Sets the quality standards to be pursued in the whole project

Exploitation plan (including IP provisions)

Describes the exploitation plan of the project including IP provisions

Manipulation platform requirements

Describes the work done in WP41 and outlines the desired specifications of the robot and vision systems based on the requirements of the manipulation tasks with the fresco pieces

Style priors and their formal description for appearance prediction

Describes the work done in WP31 where highlevel knowledge of fresco styles will be formalised in ways that allow its integration into the puzzle solving algorithm

View selection for visual perception tasks

Describes the work done in WP5.3 which will develop optimisation methods to select the arm pose that provides the most informative visual input for a successful grasp.

End-effectors modules

Describes the work done in WP 4.3 where we will leverage on soft robotic principles together with under-actuation to develop soft and adaptive end-effectors.

Strategic communication and dissemination plan

Describes the strategic communication and dissemination plan of the project

Open Research Data Pilot (1)

Data Management Plan

It provides indications as to the management of data in relation to the H2020 Open Research Data Pilot Data Management Plan might be updated during the project implementation

Websites, patent fillings, videos etc. (1)

Project website and social media (including YouTube channel)

Website of the project and social media platforms

Pubblicazioni

Conference proceedings (5)

3DSGrasp: 3D Shape-Completion for Robotic Grasp

Autori: Seyed S. Mohammadi, Nuno F. Duarte, Dimitrios Dimou, Yiming Wang, Matteo Taiana, Pietro Morerio, Atabak Dehban, Plinio Moreno, Alexandre Bernardino, Alessio Del Bue, José Santos-Victor
Pubblicato in: 2023 IEEE International Conference on Robotics and Automation (ICRA), 2024, Pagina/e 3815-3822
Editore: IEEE
DOI: 10.1109/icra48891.2023.10160350

Learning Goal-Oriented Non-Prehensile Pushing in Cluttered Scenes

Autori: Nils Dengler, David Großklaus, Maren Bennewitz
Pubblicato in: 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024, Pagina/e 1116-1122
Editore: IEEE
DOI: 10.1109/iros47612.2022.9981873

Relaxation labeling meets GANs: Solving jigsaw puzzles with missing borders

Autori: M. Khoroshiltseva, A. Traviglia, M. Pelillo and S. Vascon
Pubblicato in: Proceedings of the 21st International Conference on Image Analysis and Processing, 2022
Editore: Springer
DOI: 10.1007/978-3-031-06433-3_3

Multi-phase relaxation labeling for square jigsaw puzzle solving

Autori: B. Vardi, M. Khoroshiltseva, A. Torcinovich, M. Pelillo, and O. Ben-Shahar
Pubblicato in: Proceedings of the 18th International Joint Conference on Computer Vision, Imaging, and Computer Graphics: Theory and Applications, 2023
Editore: SCITEPRESS
DOI: 10.5220/0011622800003417

GANzzle: Reframing jigsaw puzzle solving as a retrieval task using a generative mental image

Autori: D. Talon, A. Del Bue, S. James
Pubblicato in: Proceedings of IEEE International Conference on Image Processing, 2022
Editore: IEEE
DOI: 10.1109/icip46576.2022.9897553