Please use this identifier to cite or link to this item:
http://dspace.cityu.edu.hk/handle/2031/5585
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Yeung, Wing Yee | en_US |
dc.date.accessioned | 2010-02-10T06:11:04Z | |
dc.date.accessioned | 2017-09-19T08:50:38Z | |
dc.date.accessioned | 2019-02-12T06:54:27Z | - |
dc.date.available | 2010-02-10T06:11:04Z | |
dc.date.available | 2017-09-19T08:50:38Z | |
dc.date.available | 2019-02-12T06:54:27Z | - |
dc.date.issued | 2008 | en_US |
dc.identifier.other | 2008csywy707 | en_US |
dc.identifier.uri | http://144.214.8.231/handle/2031/5585 | - |
dc.description.abstract | When people want to buy a flat, the flat may not be always available for the buyers to visit the flat in person. So, it is necessary for them to see the floor plan of the desired flat. However it is not realistic by just seeing the 2D floor plan; and it is difficult for people to imagine the actual environment. Therefore 3D model of the flat can be used. It is found that 3D modeling of a flat is not commonly used by normal users nowadays. The main reason is the common existing 3D visualization tools are difficult for non-technical users to create a 3D model of a flat with interior design. Most of the tools do not have a graphics recognition feature to automate the 3D model generation. The aim of this project is to implement a system that can import a 2D floor plan and automatically generate a 3D model according to the 2D floor plan imported. This software mainly targets on the users who do not have technical knowledge of making 3D computer graphics. So a user-friendly interface is designed for them to create a 3D model easily. In order to automate the generation of 3D model, the technique of graphics recognition is used. The objects in the 2D floor plan, including walls, doors, sofa and tables can be recognized by the system. Then object mapping is preformed. The 3D corresponding models will be transformed to map with the recognized position in the 2D floor plan. Some sample furniture are pre-loaded to the system for users to add and modify, so normal users can do the interior design by using this system. Throughout the project, techniques of graphics recognition and mapping of 3D model to 2D floor plan has been examined. In order to check the algorithm used for graphics recognition and mapping of 3D model, 60 test cases with more than 150 objects’ recognition have been done. The results showed that the overall correct rate of object recognition is as high as 93% and the average error rate of 3D models generation is low as 8%. This proved that this system has very good performance. The project would be useful for the development of the automated 3D visualization tool. The automation of 3D model generation from 2D floor plan would be the main easy-to-use feature to attract normal users. Thus, this project would be a potential product in the market. | en_US |
dc.rights | This work is protected by copyright. Reproduction or distribution of the work in any format is prohibited without written permission of the copyright owner. | en_US |
dc.rights | Access is restricted to CityU users. | en_US |
dc.title | Creation of 3D model from 2D floor plan | en_US |
dc.contributor.department | Department of Computer Science | en_US |
dc.description.supervisor | Supervisor: Dr. Wong, Raymond Hau San; First Reader: Dr. Xue, Jason Chun; Second Reader: Prof. Ip, Horace Ho Shing | en_US |
Appears in Collections: | Computer Science - Undergraduate Final Year Projects |
Files in This Item:
File | Size | Format | |
---|---|---|---|
fulltext.html | 146 B | HTML | View/Open |
Items in Digital CityU Collections are protected by copyright, with all rights reserved, unless otherwise indicated.