This tutorial is presented at a workshop in conjunction with the i3ce 2024 conference in Pittsburgh, Pennsylvania (July 28, 2024)
The past few years have seen giant leaps in machine learning, deep learning, and generative AI tools pioneered by the AI community. These tools have tremendous potential to accelerate and automate current workflows in the architecture, engineering, and construction industry related to processing spatial information in scans of the built environment. The main objective of this workshop is to provide a high-level introduction to these deep learning tools along with their applications in diverse areas such as building lifecycle management, construction monitoring, energy modeling, maintenance and renovation.
In this workshop, the participants will experience a hands-on tutorial on using open-source software tools to extract semantic information and model objects in 3D point clouds. The participants will use provided functions in CloudCompare, Python, Open3D, and PyTorch to implement tasks such as object recognition and semantic modeling of building elements. The tutorial will provide step-by-step demos on how to implement various point clouds processing algorithms such as classification and segmentation in PyTorch. The workshop will end with an open-ended discussion on recent trends, application areas and future directions for this research area.
)
NumPy Library
NumPy is an open-source Python library for numerical computing. Documentation for NumPy can be found here.
Open3D Library
Open3D is an open-source library that supports rapid development of software that deals with 3D data. Documentation for using Open3D can be found here.
PyTorch Library
PyTorch is an open-source deep learning library that is widely used for computer vision and natural language processing. PyTorch tutorials can be found here.
CloudCompare viewer
CloudCompare is an open-source 3D point cloud processing and visualization software. CloudCompare supports many open point cloud formats (ASCII, LAS, E57, etc.) as well as triangular meshes (OBJ, PLY, STL, FBX, etc.). The intermediate outputs from this notebook can be viewed and processed in CloudCompare.