Cumincad : CUMINCAD Papers : Search Results

CumInCAD is a Cumulative Index about publications in Computer Aided Architectural Design
supported by the sibling associations ACADIA, CAADRIA, eCAADe, SIGraDi, ASCAAD and CAAD futures

PDF papers
References

Hits 1 to 4 of 4

Reformat results as: short short into frame detailed detailed into frame

_id	acadia18_156
id	acadia18_156
authors	Huang, Weixin; Zheng, Hao
year	2018
title	Architectural Drawings Recognition and Generation through Machine Learning
doi	https://doi.org/10.52842/conf.acadia.2018.156
source	ACADIA // 2018: Recalibration. On imprecisionand infidelity. [Proceedings of the 38th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-692-17729-7] Mexico City, Mexico 18-20 October, 2018, pp. 156-165
summary	With the development of information technology, the ideas of programming and mass calculation were introduced into the design field, resulting in the growth of computer- aided design. With the idea of designing by data, we began to manipulate data directly, and interpret data through design works. Machine Learning as a decision making tool has been widely used in many fields. It can be used to analyze large amounts of data and predict future changes. Generative Adversarial Network (GAN) is a model framework in machine learning. It’s specially designed to learn and generate output data with similar or identical characteristics. Pix2pixHD is a modified version of GAN that learns image data in pairs and generates new images based on the input. The author applied pix2pixHD in recognizing and generating architectural drawings, marking rooms with different colors and then generating apartment plans through two convolutional neural networks. Next, in order to understand how these networks work, the author analyzed their framework, and provided an explanation of the three working principles of the networks, convolution layer, residual network layer and deconvolution layer. Lastly, in order to visualize the networks in architectural drawings, the author derived data from different layer and different training epochs, and visualized the findings as gray scale images. It was found that the features of the architectural plan drawings have been gradually learned and stored as parameters in the networks. As the networks get deeper and the training epoch increases, the features in the graph become more concise and clearer. This phenomenon may be inspiring in understanding the designing behavior of humans.
keywords	full paper, design study, generative design, ai + machine learning, ai & machine learning
series	ACADIA
type	paper
email
last changed	2022/06/07 07:49

_id	caadria2018_033
id	caadria2018_033
authors	Bai, Nan and Huang, Weixin
year	2018
title	Quantitative Analysis on Architects Using Culturomics - Pattern Study of Prizker Winners Based on Google N-gram Data
doi	https://doi.org/10.52842/conf.caadria.2018.2.257
source	T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 2, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 257-266
summary	Quantitative studies using the corpus Google Ngram, namely Culturomics, have been analyzing the implicit patterns of culture changes. Being the top-standard prize in the field of Architecture since 1979, the Pritzker Prize has been increasingly diversified in the recent years. This study intends to reveal the implicit pattern of Pritzker Winners using the method of Culturomics, based on the corpus of Google Ngram to reveal the relationship of the sign of their fame and the fact of prize-winning. 48 architects including 32 awarded and 16 promising are analyzed in the printed corpus of English language between 1900 and 2008. Multiple regression models and multiple imputation methods are used during the data processing. Self-Organizing Map is used to define clusters among the awarded and promising architects. Six main clusters are detected, forming a 3×2 network of fame patterns. Most promising architects can be told from the clustering, according to their similarity to the more typical prize winners. The method of Culturomics could expand the sight of architecture study, giving more possibilities to reveal the implicit patterns of the existing empirical world.
keywords	Culturomics; Google Ngram; Pritzker Prize; Fame Pattern; Self-Organizing Map
series	CAADRIA
email
last changed	2022/06/07 07:54

_id	caadria2018_018
id	caadria2018_018
authors	Lin, Yuming and Huang, Weixin
year	2018
title	Social Behavior Analysis in Innovation Incubator Based on Wi-Fi Data - A Case Study on Yan Jing Lane Community
doi	https://doi.org/10.52842/conf.caadria.2018.2.197
source	T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 2, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 197-206
summary	Innovation incubator is an emerging kind of office space which focuses on promoting social interaction in the space. From the perspective of environmental behavior, the complex relationship between a particular space form and the social interactions is well worth exploring. Based on Wi-Fi positioning data, this paper examined the spatial and temporal behavior in innovation incubators. Using the interdisciplinary social networks analysis, this paper further analyzed the social interactions in this space, mining out social structures such as gathering and community, and analyzing the relationship between these structures and spaces. The result shows that human behavior in innovation incubators has some interesting characteristics, and the social structures are closely linked with the functional area of innovation incubator. This paper provides a new perspective and introduces interdisciplinary approaches to study the social behaviors in a particular space form, which has great potential in future research.
keywords	environmental behavior study; social behavior analysis; innovation incubator; Wi-Fi IPS; social network
series	CAADRIA
email
last changed	2022/06/07 07:59

_id	caadria2018_049
id	caadria2018_049
authors	Xu, Tongda, Wang, Dinglu, Yang, Mingyan, You, Xiaohui and Huang, Weixin
year	2018
title	An Evolving Built Environment Prototype - A Prototype of Adaptive Built Environment Interacting with Electroencephalogram Supported by Reinforcement Learning
doi	https://doi.org/10.52842/conf.caadria.2018.2.207
source	T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 2, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 207-215
summary	This paper proposes an environment prototype learning from people's Electroencephalogram (EEG) feedback in real-time. Instead of the widely adopted supervised learning method, a recently published affordable reinforcement learning model (PPO) is adopted to avoid bias from designers and to base the interaction on the subject and intelligent agent rather than between the designer and subject. In this way, development of interaction method towards a specific target is substantially accelerated. The target of this prototype is to keep the subject's alpha wave stable or decline, which indicated a more calming state, by intelligent decision of illumination state according to subject's EEG. The result is promising, a decent trained model could be gained within 500,000 steps facing this mid-complex environment. The target of keeping the alpha wave of subjects on a low or stable level purely by decision from computer agents is successfully reached.
keywords	Brain–computer interface; Reinforcement learning; Adaptive environment; Electroencephalogram; Mindfulness training
series	CAADRIA
email
last changed	2022/06/07 07:57

No more hits.