Architectural Factors Detection from Plan by Deep Learning Framework

被引:4
作者
Park, Jinho [1 ]
Yoon, Doyoung [1 ]
Lee, Woo-Hyoung [2 ]
机构
[1] Soongsil Univ, Global Sch Media, Seoul 156743, South Korea
[2] Namseoul Univ, Dept Architecture, Cheonan 331707, South Korea
来源
INTERNATIONAL JOURNAL OF GRID AND DISTRIBUTED COMPUTING | 2018年 / 11卷 / 01期
基金
新加坡国家研究基金会;
关键词
aesthetic value; deep learning; object detection; convolutional neural net;
D O I
10.14257/ijgdc.2018.11.1.06
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
This study presents the basics of emotion - based technology system by detecting various elements of architectural drawings as the first step of objectification of aesthetic characteristic and design tendency reflecting architect 's design intention by using example - based machine learning. With the development of technology, new software has been introduced in the field of architectural design, which is able to solve the concerns of environmental values and solve problems, and generalize environmental analysis. However, the aesthetic value compatible with it does not meet the needs of the same context due to social recognition and technical limitations. The purpose of this study is to quantify the scope of subjective judgment of architectural design and to objectify it. In other words, the intention of the designer is the process of confirming the inherent attributes of the result of the expressed architectural design.
引用
收藏
页码:57 / 64
页数:8
相关论文
共 8 条
[1]  
Gatys Leon A., P IEEE C COMP VIS PA
[2]  
Jaitly N, 2012, 13TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2012 (INTERSPEECH 2012), VOLS 1-3, P2577
[3]  
Lee W.-H., 2017, ADV SCI TECHNOLOGY L, V148, P11
[4]  
Liu W, 2016, **DROPPED REF**, DOI DOI 10.1007/978-3-319-46448-0_2
[5]  
McLane Y., 2015, P 10 INT SPAC SYNT S
[6]  
Redmon J., 2016, P IEEE C COMP VIS PA, P779
[7]  
Ren S., 2015, ADV NEURAL INF PROCE, P91
[8]  
Simonyan K., 2015, ICLR 2015 INT C LEAR