Spare Parts 3D is to lead a three-year R&D programme that will see the company reconstruct a 3D model from a 2D drawing for the French Defense Innovation Agency.
The programme is called RAPID and will be carried out in partnership with the University for Research in Automated Production (LURPA) of the École Normale Supérieure Paris-Saclay, starting March 15, 2022.
Through this R&D effort, Spare Parts 3D will automate the process of 3D model reconstruction by combining deep learning and computer vision techniques with existing semantic recognition and Optical Character Recognition capabilities. The objective is to develop the application depicted in the 2D drawing and make it possible to automatically reconstitute a 3D file usable for 3D printing.
Spare Parts 3D will utilise the DigiPART software, which allows the creation of digital inventories and is able to carry out technical analysis with partial data. Using complex algorithms, DigiPART is able to combine semantic recognition, reading 2D plans, with material databases and additive manufacturing processes. The company will deploy this software during the RAPID project in a bid to create an additive manufacturing solution – comprising a 3D model, the printed part and the cost information – from a 2D drawing and a part description.
“The reconstruction of 3D models from technical drawings is a real challenge for which the combination of digital geometry processing techniques, expert and deep learning algorithms will pave the way for considerable breakthrough in the creation of digital inventories,” commented Nabil Anwer, Deputy Director of LURPA.
“Thanks to this programme, we have set up a team at the forefront of technology for the development of this fundamental software module,” added Christophe Migliorini, CTO of Spare Parts 3D. “This will strengthen the attractiveness of our DigiPART solution and consolidate our leadership in the identification of parts for additive manufacturing.”