@inproceedings{simbuild2024_2287,
	doi = {},
	url = {https://publications.ibpsa.org/conference/paper/?id=simbuild2024_2287},
	year = {2024},
	month = {May},
	publisher = {IBPSA-USA},
	author = {SoumyaDeep Chowdhury and  Kuljeet Singh Grewal},
	title  = {Rapid Building Feature Extraction and Geometry Formulation Using Machine Learning},
	booktitle = {Proceedings of SimBuild Conference 2024},
	volume  = {11},
	isbn = {},
	address  = {Denver, Colorado},
	series  = {IBPSA-USA Building Simulation Conference},
	pages = {429--440},
	abstract = {Increasing energy consumption has led to wide-scale optimization efforts of urban and sub-urban environments. To supplement this, building-energy-modeling (BEM) methods are applied to provide insights and identify optimizations. Several inputs are required for BEM, including climate, usage, and most importantly geometric data. Creation of geometric data is a time-consuming endeavor. Methods of automation often incorporate expensive or not widely available technology such as light detection and ranging (LiDAR) or unmanned aerial vehicle (UAV) imagery. This document aims to provide and prove the viability of a methodology to create building models at high levels-of-detail (LOD), using only readily available sources such as OpenStreetMaps (OSM) and street-view images (SVIs). Modern image processing techniques and machine learning algorithms such as convolutional-neural-networks (CNNs) and regional-convolutional-neural networks (R-CNNs) are explored with the goal of creating building geometry model for energy modeling with machine learning algorithms reaching a precision of 71%, with geometry creation reaching an average accuracy of 95% compared to models made from on-site manual measurements. The process can easily be extended to be user-assisted to increase overall accuracy and reduce the time complexity of the workflow for building geometry creation outputting point-cloud data which will be ultimately applied for BEM.},
	issn = {},
	Organisation = {IBPSA-USA},
	Editors = {}
}