manufacturing
AI-PRISM: AI Powered human-centred Robot Interactions for Smart Manufacturing
The goal of the project AI-PRISM is to deliver a human-centered AI-based solutions ecosystem (with and without a physical embodiment) targeted to manufacturing scenarios with tasks difficult to automate (requiring complex perception and manipulation operations) and in which speed and versatility are essential to satisfy customer demand. The result will be an integrated, scalable and yet installation-specific solution for semi-automated and collaborative manufacturing in flexible production processes. The project will integrate and adapt novel AI approaches, methods, and techniques to enable collaborative robots, AGV and automation systems to acquire optimised perception and cognitive capabilities upon interaction with environment, products, other systems and humans, within changing dynamic and partially unpredictable manufacturing scenarios.
Partners
consortium coordinated by NTT DATA SPAIN, with 3 partners and 22 participants from 12 countries, including Fondazione Bruno Kessler.
Grant
Grant agreement No:
Call: HORIZON-CL4-2021-TWIN-TRANSITION-01
Type of action: HORIZON-IA
Start date: 01/08/2022
Duration: 36 Months
Kick Off - 4 Oct 2022
TeV publications
A. Caraffa, D. Boscaini, A. Hamza, F. Poiesi. FreeZe: Training-free zero-shot 6D pose estimation with geometric and vision foundation models. European Conference on Computer Vision - ECCV, September 2024
M. Bortolon, T. Tsesmelis, S. James, F. Poiesi, A. Del Bue. 6DGS: 6D pose estimation from a single image and a 3D Gaussian Splatting model. European Conference on Computer Vision - ECCV, September 2024
L. Riz, C. Saltory, Y. Wang, E. Ricci, F. Poiesi. Novel class discovery meets foundation models for 3D semantic segmentation. International Journal on Computer Vision - IJCV, 2024
J. Corsetti, D. Boscaini, C. Oh, A. Cavallaro, F. Poiesi. Open-vocabulary object 6D pose estimation. IEEE Conference on Computer Vision and Pattern Recognition - CVPR, June 2024
M. Bortolon, T. Tsesmalis, S. James, F. Poiesi, A. Del Bue. IFFNeRF: Initialization Free and Fast 6DoF pose estimation from a single image and a NeRF model. International Conference on Robotics and Automation - ICRA, May 2024.
J. Corsetti, D. Boscaini and F. Poiesi. Revisiting Fully Convolutional Geometric Features for Object 6D Pose Estimation. International Conference on Computer Vision Workshops, October 2023
L. Riz, C. Saltori, E. Ricci and F. Poiesi. Novel class discovery for 3D point cloud semantic segmentation. IEEE Conference on Computer Vision and Pattern Recognition - CVPR, June 2023