Auto-conversion from2D drawing to 3D model with deep learning

Download

index.pdf

Date

2019

Author

Yetiş, Gizem

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

223
views

542
downloads

Modeling has always been important as it transfers knowledge to end users. From the very first line on a computer screen to AR/VR applications have broaden the perception, communication and implementation of design-related industries, and each representation technique has become one another's base, information source and data supplier. Yet, transforming the information that one includes into another has still major problems. It requires precise data, qualified personnel and human intervention. This research aims to represent an automated reconstruction from low level data sources to higher level digital models in order to eliminate these problems. This autoconversion process only examines the architectural usage and makes a sample of its usability in different fields. 2D floor plans and elevation drawings in raster format, which are collected and/or produced from scratch, are used as datasets. These drawings are semantically segmented with three different Convolutional Neural Networks to obtain relevant architectural information since Deep Learning shows promising success to solve a wide range of problem with its widespread use. Semantically segmented drawings are then transformed into 3D by using Digital Geometry Processing methods. Lastly, a web application is introduced to allow any user to obtain a 3D model with ease. Semantic segmentation results in 2D and two case studies in 3D are evaluated and compared separately with different metrics to represent accuracy of the process. To conclude, this research has proposed an automated process for reconstruction of 3D models with the state-of-the-art methods and made it ready for use even for a person without technical knowledge.

Subject Keywords

Architectural drawing., Keywords: Architectural Drawing Dataset, 3D Reconstruction, Semantic Segmentation, Convolutional Neural Networks, Digital Geometry Processing.

URI

http://etd.lib.metu.edu.tr/upload/12624598/index.pdf
https://hdl.handle.net/11511/44262

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

An automated quality measurement approach for business process models Gürbüz, Özge; Demirörs, Onur; Güceğlioğlu, A. Selçuk; Department of Information Systems (2011) Business process modeling has become a common need for organizations. Therefore process quality is also having an important role for the organizations. The most of the quality studies are based on cost and time which can be analyzed during or after the execution of the business processes. There are also quality measures which help analyzing measures before the execution of the business processes. This type of measures can give early feedback about the processes. There are three frameworks defined in the lit...
Computer Simulation as an Integral Part of Digital Design Education in Architecture The Modulator Sorguç, Arzu; Cakici, Fatma Zehra (2011-09-24) It has been more than a decade that the first discussions on the integration of computational technologies into architectural education, especially into the design studios have been started. Since then, the use of CAD from drafting to complex solid modeling and manufacturing in design studios has been mostly achieved. Yet, persisting on the discussions within the same context, mostly focusing on the studio education and CAD/CAM rather than "CAAD" avoids inquiring other potentials of them as being a learning...
Characterizing the development and usage of diagrams in embedded software systems Akdur, Deniz; Demirörs, Onur; Garousi, Vahid (2017-09-01) To cope with growing complexity of embedded software, modeling has become popular. The usage of models in embedded software industry and the relevant practices usually vary since the purposes of diagram development and usage differ. Since a large variety of software modeling practices used in embedded software industry, it is important to understand its state-of-the-practice and its usage degree while investigating the relations between its attributes (e.g., modeling rigor, purpose, code correspondence, sta...
Software subcontracting success: a case study on the relationship betweeen project success and process metrics Yücetürk, Kerem; Bilgen, Semih; Department of Information Systems (2005) While software subcontracting is a common business transaction in the information technology field, metrics specifically aimed at effectiveness of software subcontracting arrangements are not commonly available. This thesis makes a review of the literature and derives such metrics from fields of software quality, COTS acquisition and IS success. A case study is performed on software subcontracting projects of a Turkish defense contractor, and the project metrics are compared according to their success. The ...
A proposal for the structure of production drawings of “design / build” construction projects in computer environment Çal, Nazlı Naz; Tanyer, Ali Murat; Department of Building Science in Architecture (2007) Documentation is an important aspect in the construction industry since there are many types and number of documents that have to be controlled for the success of a given project. In design/build projects, site technical groups of the contractor upgrade the tender drawings by preparing shop drawings to be sent to the consultant for checking purpose. Especially in large projects a great many shop drawings are produced, causing a need of a system to keep track of and to understand the status of the drawings. ...

Citation Formats

G. Yetiş, “Auto-conversion from2D drawing to 3D model with deep learning,” Thesis (M.S.) -- Graduate School of Natural and Applied Sciences. Building Science in Architecture., Middle East Technical University, 2019.