Enhancing 3D Clay Printing with Computer Vision and Deep Learning
Serdar Așut (TU Delft - Architecture and the Built Environment)
X. Ding (TU Delft - Architecture and the Built Environment)
S. Guha (Student TU Delft)
S. Ryu (Student TU Delft)
W. Wei (Student TU Delft)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
This paper presents an ongoing project that explores the usability of computer vision and deep learning to improve the quality of 3D clay printing (3DCP). One of the challenges in 3DCP is related to the nonstandard nature of the clay mixture and the environmental conditions in which the printing happens, which can result in printing failures. Manual interventions are required to adjust the printing parameters to ensure a good result. In this project, we aimed to develop an automated solution to this challenge by using computer vision and the Attention-56 deep learning network (DLN) method presented by Wang et al. (2017) and the real-time material flow control method presented by Brion and Pattinson (2022a, 2022b). Our work adapts these methods for 3DCP to adjust layer height and extrusion amount to automatically respond to changing clay mixture properties and achieve better results.