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 20 of 340

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

_id	cdrf2022_223
id	cdrf2022_223
authors	Zhiyi Dou, Waishan Qiu, Wenjing Li, Dan Luo
year	2022
title	Evaluation Process of Urban Spatial Quality and Utility Trade-Off for Post-COVID Working Preferences
doi	https://doi.org/https://doi.org/10.1007/978-981-19-8637-6_19
source	Proceedings of the 2022 DigitalFUTURES The 4st International Conference on Computational Design and Robotic Fabrication (CDRF 2022)
summary	The formation of cities, and the relocation of workers to densely populated areas reflect a spatial equilibrium, in which the higher real consumption levels of urban areas are offset by lower non-monetary amenities [1]. However, as the society progress toward a post-COVID stage, the prevailing decentralized delivery systems and location-based services, the growing trend of working from home, with citizens’ shifting preference of de-appreciating densities and gathering, have not only changed the possible spatial distribution of opportunities, resources, consumption and amenities, but also transformed people’s preference regarding desirable urban spatial qualities, value of amenities, and working opportunities [2, 3]. This research presents a systematic method to evaluate the perceived trade-off between urban spatial qualities and urban utilities such as amenities, transportation, and monetary opportunities by urban residence in the post-COVID society. The outcome of the research will become a valid tool to drive and evaluate urban design strategies based on the potential self-organization of work-life patterns and social profiles in the designated neighbourhood. To evaluate the subjective perception of the urban residence, the study started with a comparative survey by asking residence to compare two randomly selected urban contexts in a data base of 398 contexts sampled across Hong Kong and state their living preference under the presumption of following scenarios: 1. working from home; 2. working in city centre offices. Core information influencing the spatial equilibrium are provided in the comparable urban context such as street views, housing price, housing space, travel time to city centre, adjacency to public transport and amenities, etc. Each context is given a preference score calculated with Microsoft TrueSkill Bayesian ranking algorithm [4] based on the comparison survey of two scenarios. The 398 contexts are further analysed via GIS and image processing, to be deconstructed into numerical values describing main features for each of the context that influence urban design strategies such as composition of spatial features, amenity allocation, adjacency to city centre and public transportations. Machine learning models are trained with the numerical values of urban features as input and two preference scores for the two working scenarios as the output. The correlation heat maps are used to identify main urban features and its p-value that influence residence’s preference under two working scenarios in post–COVID era. The same model could also be applied to inform the direction of urban design strategies to construct a sustainable community for each type of working population and validate the design strategies via predicting its competitiveness in attracting residence and developing target industries.
series	cdrf
email
last changed	2024/05/29 14:02

_id	caadria2022_90
id	caadria2022_90
authors	Veloso, Pedro, Rhee, Jinmo, Bidgoli, Ardavan and Ladron de Guevara, Manuel
year	2022
title	Bubble2Floor: A Pedagogical Experience With Deep Learning for Floor Plan Generation
doi	https://doi.org/10.52842/conf.caadria.2022.1.373
source	Jeroen van Ameijde, Nicole Gardner, Kyung Hoon Hyun, Dan Luo, Urvi Sheth (eds.), POST-CARBON - Proceedings of the 27th CAADRIA Conference, Sydney, 9-15 April 2022, pp. 373-382
summary	This paper reports a pedagogical experience that incorporates deep learning to design in the context of a recently created course at the Carnegie Mellon University School of Architecture. It analyses an exercise called Bubble2Floor (B2F), where students design floor plans for a multi-story row-house complex. The pipeline for B2F includes a parametric workflow to synthesise an image dataset with pairs of apartment floor plans and corresponding bubble diagrams, a modified Pix2Pix model that maps bubble diagrams to floor plan diagrams, and a computer vision workflow to translate images to the geometric model. In this pedagogical research, we provide a series of observations on challenges faced by students and how they customised different elements of B2F, to address their personal preferences and problem constraints of the housing complex as well as the obstacles from the computational workflow. Based on these observations, we conclude by emphasising the importance of training architects to be active agents in the creation of deep learning workflows and make them accessible for socially relevant and constrained design problems, such as housing.
keywords	Architectural Pedagogy, Deep Learning, Conditional GAN, Space Planning, Floor Plan, SDG 4, SDG 9
series	CAADRIA
email
last changed	2022/07/22 07:34

_id	caadria2022_42
id	caadria2022_42
authors	Chen, Jielin and Stouffs, Rudi
year	2022
title	Robust Attributed Adjacency Graph Extraction Using Floor Plan Images
doi	https://doi.org/10.52842/conf.caadria.2022.2.385
source	Jeroen van Ameijde, Nicole Gardner, Kyung Hoon Hyun, Dan Luo, Urvi Sheth (eds.), POST-CARBON - Proceedings of the 27th CAADRIA Conference, Sydney, 9-15 April 2022, pp. 385-394
summary	Architectural design solutions are intrinsically structured information with a broad range of interdependent scopes. Compared to conventional 2D Euclidean data such as orthographic drawings and perspectives, non-Euclidean data (e.g., attributed adjacency graphs) can be more effective and accurate for representing 3D architectural design information, which can be useful for numerous design tasks such as spatial analysis and reasoning, and practical applications such as floor plan parsing and generation. Thus, getting access to a matching attributed adjacency graph dataset of architectural design becomes a necessity. However, the task of conveniently acquiring attributed adjacency graphs from existing architectural design solutions still remains an open challenge. To this end, this project leverages state-of-the-art image segmentation techniques using an ensemble learning scheme and proposes an end-to-end framework to efficiently extract attributed adjacency graphs from floor plan images with diverse styles and varied levels of complexity, aiming at addressing generalization issues of existing approaches. The proposed graph extraction framework can be used as an innovative tool for advancing design research infrastructure, with which we construct a large-scale attributed adjacency graph dataset of architectural design using floor plan images retrieved in bulk. We have open sourced our code and dataset.
keywords	attributed adjacency graph, floor plan segmentation, ensemble learning, architectural dataset, SDG 9
series	CAADRIA
email
last changed	2022/07/22 07:34

_id	cdrf2022_488
id	cdrf2022_488
authors	Tomás Vivanco, Juan Eduardo Ojeda, Philip Yuan
year	2022
title	Regression-Based Inductive Reconstruction of Shell Auxetic Structures
doi	https://doi.org/https://doi.org/10.1007/978-981-19-8637-6_42
source	Proceedings of the 2022 DigitalFUTURES The 4st International Conference on Computational Design and Robotic Fabrication (CDRF 2022)
summary	This article presents the design process for generating a shell-like structure from an activated bent auxetic surface through an inductive process based on applying deep learning algorithms to predict a numeric value of geometrical features. The process developed under the Material Intelligence Workflow applied to the development of (1) a computational simulation of the mechanical and physical behaviour of an activated auxetic surface, (2) the generation of a geometrical dataset composed of six geometric features with 3,000 values each, (3) the construction and training of a regression Deep Neuronal Network (DNN) model, (4) the prediction of the geometric feature of the auxetic surface's pattern distance, and (5) the reconstruction of a new shell based on the predicted value. This process consistently reduces the computational power and simulation time to produce digital prototypes by integrating AI-based algorithms into material computation design processes.
series	cdrf
email
last changed	2024/05/29 14:03

_id	sigradi2022_6
id	sigradi2022_6
authors	Abu-Aridah, Dima; Ligler, Heather
year	2022
title	From Shelter to Home: Transformation Grammar of Housing Units in Irbid Refugee Camp
source	Herrera, PC, Dreifuss-Serrano, C, Gómez, P, Arris-Calderon, LF, Critical Appropriations - Proceedings of the XXVI Conference of the Iberoamerican Society of Digital Graphics (SIGraDi 2022), Universidad Peruana de Ciencias Aplicadas, Lima, 7-11 November 2022 , pp. 311–322
summary	This paper presents research on the design challenges in refugee camps where “temporary" shelters often evolve into permanent homes and larger communities. These transformations convey an informal design process, a phenomenon evident in Irbid Camp for Palestinian refugees in Jordan. To study this site and design process in detail, shape rules based on the transformation of ten individual housing units are developed, with consideration of area and growth limitations inside the refugee camp. The Irbid Camp Grammar reveals a modular, grid-based logic at play in the incremental and spontaneous design of refugee housing from temporary shelters to permanent homes. This study is one step forward in helping us understand how formalizing this growth logic can contribute to the design of better emergency housing interventions in the future.
keywords	Shape grammars, Emergency housing, Refugee housing, Housing transformation, Informal settlements
series	SIGraDi
email
last changed	2023/05/16 16:55

_id	sigradi2022_104
id	sigradi2022_104
authors	Bielski, Jessica; Eisenstadt, Viktor; Langenhan, Christoph; Petzold, Frank
year	2022
title	Lost in architectural designing - Possible cognitive biases of architects during the early design phases
source	Herrera, PC, Dreifuss-Serrano, C, Gómez, P, Arris-Calderon, LF, Critical Appropriations - Proceedings of the XXVI Conference of the Iberoamerican Society of Digital Graphics (SIGraDi 2022), Universidad Peruana de Ciencias Aplicadas, Lima, 7-11 November 2022 , pp. 557–568
summary	In order to meet the housing demands of the future, architects need to work faster and more efficiently while improving architectural quality. The metis projects aim to create an intelligent design assistant supporting architects during the early design stages through suggesting further design steps for spatial layouting, based on the best practice of reference buildings. By enhancing suggestions with explainability, the system offers insight to improve Human-System-Interaction (HSI), bridging the ‘black box’ problem. The explanations aim to either support the reasoning process or mitigate possible biases of architects, which can be rooted in the heuristic ‘System 1’, as well as the analytical ‘System 2’, drawing from the ‘dual process model’. Within this paper, we propose our approach to clarify the four main heuristic biases and the logical errors of architects, when using reference buildings, and their respective representation during the architectural design decision-making process.
keywords	Decision Making, Biases, Explainability, XAI, Human System Interaction
series	SIGraDi
email
last changed	2023/05/16 16:56

_id	ecaade2022_411
id	ecaade2022_411
authors	Cesar Rodrigues, Ricardo, Rubio Koga, Renan, Hitomi Hirota, Ercilia and Bertola Duarte, Rovenir
year	2022
title	Mapping Space Allocation with Artificial Intelligence - An approach towards mass customized housing units
doi	https://doi.org/10.52842/conf.ecaade.2022.2.631
source	Pak, B, Wurzer, G and Stouffs, R (eds.), Co-creating the Future: Inclusion in and through Design - Proceedings of the 40th Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2022) - Volume 2, Ghent, 13-16 September 2022, pp. 631–640
summary	Artificial Intelligence represents a substantial part of the available tools on architectural design, especially for Space Layout Planning (SLP). At the same time, the challenge of Mass Customization (MC) is to increase the product variety while maintaining a good cost-benefit ratio. Thus, this research aims to identify new, valid, and easily understandable data patterns through human-machine interaction in an attempt to deal with the challenges of MC during the early phases of SLP. The Design Science Research method was adopted to develop a digital artifact based on deep generative models and a reverse image search engine. The results indicate that the artifact can deliver a series of design alternatives and enhance the navigation process in the solution space, besides giving key insights on dataset design for further research.
keywords	Floor plans, Generative Adversarial Networks, Mass Customization
series	eCAADe
email
last changed	2024/04/22 07:10

_id	sigradi2022_15
id	sigradi2022_15
authors	Jiang, Wanzhu; Wang, Jiaqi
year	2022
title	Autonomous Collective Housing Platform: Digitization, Fluidization and Materialization of Ownership
source	Herrera, PC, Dreifuss-Serrano, C, Gómez, P, Arris-Calderon, LF, Critical Appropriations - Proceedings of the XXVI Conference of the Iberoamerican Society of Digital Graphics (SIGraDi 2022), Universidad Peruana de Ciencias Aplicadas, Lima, 7-11 November 2022 , pp. 15–26
summary	New social phenomena like digital nomads urge an upgrade in housing ownership. This research proposes an autonomous housing platform that shapes residential communities into adaptive and reconfigurable systems, framing a cycle of digitalization, fluidization and materialization of housing ownership. Specifically, the interactive interface carries the flexible ownership model that uses virtual space voxels as digital currency; the artificial intelligence algorithm drives the multilateral ownership negotiation and circulation, and modular robots complete the mapping from ownership status to real spaces. Taking project TESSERACT as a case study, we verified the feasibility of this method and presented expected co-living scenarios: the spaces and ownership are constantly adjusted according to demands and are always in the closest interaction with users. By exploring the ownership evolution, this research guides an integrated and inclusive housing system paradigm, triggering critical evaluation of traditional models and providing new ideas for solving housing problems in the post-digital era.
keywords	Agent-Based Systems, Digital Platform, Housing Ownership, Space Planning Algorithm, Discrete Material System
series	SIGraDi
email
last changed	2023/05/16 16:55

_id	caadria2022_344
id	caadria2022_344
authors	Krezlik, Adrian
year	2022
title	Considering Energy, Materials and Health Factors in Architectural Design, Two Renovation Strategies for the Portuguese Building Stock
doi	https://doi.org/10.52842/conf.caadria.2022.2.619
source	Jeroen van Ameijde, Nicole Gardner, Kyung Hoon Hyun, Dan Luo, Urvi Sheth (eds.), POST-CARBON - Proceedings of the 27th CAADRIA Conference, Sydney, 9-15 April 2022, pp. 619-628
summary	According to the Intergovernmental Panel on Climate Change, the built environment has a significant share in global final energy use, greenhouse gases emission, land-system change, and biodiversity loss to list some indicators. In Europe, the biggest challenge is to regenerate existing building stock to create a positive impact on Nature. The Portuguese housing stock is old: 56% is more than 30 years old, and it has a low level of thermal comfort and energy efficiency. The first thermal regulations appeared in 1990 and therefore most of the houses need urgent renovation to meet EU decarbonization goals, and to improve energy efficiency, as well as well-being and comfort of residents. This paper presents a method that aims to verify existing solutions known from vernacular architecture as complementary to existing strategies. It employs digital simulation to verify whether they could be used for renovation, measuring their impact on human and planetary health. The paper shows that there is a wide spectrum of parameters that influence the renovation process and that it is possible to enhance building performance using vernacular knowledge.
keywords	Building Energy Modelling, Life Cycle Assessment, Occupant Health, Energy Renovation, Vernacular Mimicry, SDG 3, SDG 11, SDG 13
series	CAADRIA
email
last changed	2022/07/22 07:34

_id	ijac202220407
id	ijac202220407
authors	Lacroix, Igor; Orkan Zeynel Güzelci; Gonçalo Furtado Lopes; José Pedro Sousa
year	2022
title	Connecting the Portuguese system of evolutive housing with building information modeling: From analogical to digital methods
source	International Journal of Architectural Computing 2022, Vol. 20 - no. 4, pp. 801–816
summary	In Portugal, in the 1960s and 1970s, there was research concerning a system of the architectural design of housing for economically less favored populations, which related sociological information with analogical computational methods and culminated with its application in the Local Ambulatory Support Service (SAAL). This article presents the digitization process of these methods for the development of an architectural design system for social housing. The main goal is to improve methodological procedures for the original research and, specifically, to adapt them to computational design and modeling processes. To this end, this research transposed the aforementioned methodology into an algorithmic model that matches sociological information acquired from an online form with a database of social housing floor plan images to generate a building information modeling (BIM) directly from the selected image source. The result is an algorithmic model informed by sociological data linked with a BIM model to enable further rationalization of architectural design.
keywords	evolutive housing, social housing, local ambulatory support service, sociological survey, algorithmic modeling, building information modeling
series	journal
last changed	2024/04/17 14:30

_id	caadria2022_77
id	caadria2022_77
authors	Marschall, Max and Sepulveda, Pablo
year	2022
title	How to Prevent a Passive House from Overheating: An Industry Case Study Using Parametric Design to Propose Compliance Strategies
doi	https://doi.org/10.52842/conf.caadria.2022.2.639
source	Jeroen van Ameijde, Nicole Gardner, Kyung Hoon Hyun, Dan Luo, Urvi Sheth (eds.), POST-CARBON - Proceedings of the 27th CAADRIA Conference, Sydney, 9-15 April 2022, pp. 639-648
summary	The airtight, well-insulated building fabric of a Passive House can reduce operational energy consumption but can also present a risk of overheating during summer. PHPP, the Excel tool used to model Passive Houses, considers the whole building as a single thermal zone; a simplification that might be partly responsible for the tool‚s limited ability to predict overheating risk. The current study on a real-world project provides insights on two topics. First, we compare PHPP‚s overheating assessment with that of CIBSE‚s TM59 standard that requires dynamic energy modelling at a room level. Our results support the claim that PHPP underestimates overheating; in our case, glazing SHGC and air change rate were some of the most important parameters affecting compliance, as were some other, rarely analysed factors like ratio of external wall to room volume. Second, we report on the effectiveness of using parametric design for compliance modelling of this kind, and found that parameter studies, coupled with appropriate data visualisation, are an effective way to build intuition on a design problem of this kind.
keywords	Passive House, social housing, EnergyPlus modelling, PHPP modelling, overheating risk, parametric data visualisation, SDG 3, SDG 13
series	CAADRIA
email
last changed	2022/07/22 07:34

_id	sigradi2022_94
id	sigradi2022_94
authors	Melo Santos Paulino, Daniele; J Knapp, Cooper; Ligler, Heather; Napolitano, Rebecca
year	2022
title	The Reviver Grammar: transforming the historic center of Sao Luís through social housing
source	Herrera, PC, Dreifuss-Serrano, C, Gómez, P, Arris-Calderon, LF, Critical Appropriations - Proceedings of the XXVI Conference of the Iberoamerican Society of Digital Graphics (SIGraDi 2022), Universidad Peruana de Ciencias Aplicadas, Lima, 7-11 November 2022 , pp. 347–358
summary	This work explores the potential of Shape Grammars in generating solutions for the adaptive reuse of historic buildings. It proposes a transformation grammar for the Sobrado building, a typology present in the historic center of Sao Luís. The methodology proposes a framework for adapting buildings into multi-family apartments, considering spatial and structural requirements. The grammar aims to develop a formalism for repurposing historic buildings into social housing and considers the allocation of three types of apartments in the floor plan: studios, one-bedroom, and two-bedroom apartments. The adopted strategy for the distribution of internal spaces considers opening elements, such as windows and balconies, allowing to benefit from the natural daylighting characteristics of the buildings. This paper describes the grammar rules and its application to a case study building, aiming to demonstrate how the grammar supports different layout solutions for the same design space.
keywords	shape grammars, adaptive reuse, generative design, mass customization, sustainability
series	SIGraDi
email
last changed	2023/05/16 16:55

_id	ascaad2022_063
id	ascaad2022_063
authors	Ozman, Gizem; Selcuk, Semra
year	2022
title	Generating Mass Housing Plans through GANs: A case in TOKI, Turkey
source	Hybrid Spaces of the Metaverse - Architecture in the Age of the Metaverse: Opportunities and Potentials [10th ASCAAD Conference Proceedings] Debbieh (Lebanon) [Virtual Conference] 12-13 October 2022, pp. 17-29
summary	Nowadays, Machine Learning (ML) is frequently used in almost all disciplines having an intersection with technology. Recently, architects are using existing plan data sets in architecture through Deep Learning (DL) algorithms of big data to achieve generative and non-existent plan models by using ML. Especially, Generative Adversarial Neural Networks (GANs), one of the deep learning algorithms, have been in use in the creation of generative models for architectural studies. Within the scope of this paper, architectural drawings were generated by using GANs. This generation method allows for the training of spatial layout planning to networks and for the generation of plans that do not exist in the dataset. Architectural drawings of TOKI (Housing Development Administration of the Republic of Türkiye) mass housing projects were used as datasets. In line with studies already carried out, this study attempts to create a method for further processing of the research. In this study, the differences between the plan typologies generated with raster images and the reality relations in visual productions between graph-based plan layout productions were evaluated. In this context, 157 plan datasets were obtained by multiplying plans which were spatially correlated with the RGB settings of 21 plan typologies. As a result of this research, it has been determined that the spatial layout planning of the HouseGAN algorithm provides TOK?'s current plan typologies of generation together with bubble diagrams. HouseGAN was trained using its dataset and the outputs obtained were realistic background images.
series	ASCAAD
email
last changed	2024/02/16 13:29

_id	sigradi2022_107
id	sigradi2022_107
authors	Velasco, Rodrigo; Luna, Leonardo; Hudson, Roland; Shepherd, Paul
year	2022
title	Concrete Construction for Social Housing in Colombia: Towards a higher productivity supported by the use of Digital tools and off-site processes.
source	Herrera, PC, Dreifuss-Serrano, C, Gómez, P, Arris-Calderon, LF, Critical Appropriations - Proceedings of the XXVI Conference of the Iberoamerican Society of Digital Graphics (SIGraDi 2022), Universidad Peruana de Ciencias Aplicadas, Lima, 7-11 November 2022 , pp. 569–580
summary	This paper presents the outcomes of a collaborative project between UK and Colombian researchers, aimed at increasing productivity and lowering emissions for the construction of social housing in Colombia, particularly when using concrete as the main building material. Based on the findings of a previous project that focused on the automation of concrete construction within the UK context, and following the implementation of interviews with local industry experts, the researchers analyzed the opportunities and challenges of incorporating digital technologies in order to improve the construction process, and found out that there was higher potential in the design and management areas rather than in the actual production of components, where standard prefabrication was perceived as the most fitting solution. This paper will introduce some of the advances in the project, which include a diagnosis of the challenges and obstacles of prefabricated systems used in construction, obtained from interviews with leaders and actors in the sector, a meta-analysis of barriers and opportunities of the use of concrete prefabrication in other contexts, the development of a typology as an example of an application developed as part of the project development, and finally a software tool to support the proposed case of application.
keywords	Concrete construction, Prefabrication, Social housing, FEA Analysis, Generative tools
series	SIGraDi
email
last changed	2023/05/16 16:56

_id	ecaade2022_398
id	ecaade2022_398
authors	Dzurilla, Dalibor and Achten, Henri
year	2022
title	What’s Happening to Architectural Sketching? - Interviewing architects about transformation from traditional to digital architectural sketching as a communicational tool with clients
doi	https://doi.org/10.52842/conf.ecaade.2022.1.389
source	Pak, B, Wurzer, G and Stouffs, R (eds.), Co-creating the Future: Inclusion in and through Design - Proceedings of the 40th Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2022) - Volume 1, Ghent, 13-16 September 2022, pp. 389–398
summary	The paper discusses 23 interviewed architects in practice about the role of traditional and digital sketching (human-computer interaction) in communication with the client. They were selected from 1995 to 2018 (the interval of graduation) from three different countries: the Czech Republic (CR), Slovakia (SR), Netherland (NR). To realize three blending areas that impact the approach to sketching: (I) Traditional hand and physical model studies (1995-2003). (II)Transition form - designing by hand and PC (2004–2017). (III) Mainly digital and remote forms of designing (2018–now). Interviews helped transform 31 “parameters of tools use” from the previous theoretical framework narrowed down into six main areas: (1) Implementation; (2)Affordability; (3)Timesaving; (4) Drawing support; (5) Representativeness; (6) Transportability. Paper discusses findings from interviewees: (A) Implementation issues are above time and price. (B) Strongly different understanding of what digital sketching is. From drawing in Google Slides by mouse to sketching in Metaverse. (C) Substantial reduction of traditional sketching (down to a total of 3% of the time) at the expense of growing responsibilities. (D) 80% of respondents do not recommend sketching in front of the client. Also, other interesting findings are further described in the discussion.
keywords	Architectural Sketch, Digital Sketch, Effective Visual Communication
series	eCAADe
email
last changed	2024/04/22 07:10

_id	caadria2022_277
id	caadria2022_277
authors	Akbar, Zuardin, Wood, Dylan, Kiesewetter, Laura, Menges, Achim and Wortmann, Thomas
year	2022
title	A Data-Driven Workflow for Modelling Self-Shaping Wood Bilayer, Utilizing Natural Material Variations with Machine Vision and Machine Learning
doi	https://doi.org/10.52842/conf.caadria.2022.1.393
source	Jeroen van Ameijde, Nicole Gardner, Kyung Hoon Hyun, Dan Luo, Urvi Sheth (eds.), POST-CARBON - Proceedings of the 27th CAADRIA Conference, Sydney, 9-15 April 2022, pp. 393-402
summary	This paper develops a workflow to train machine learning (ML) models with a small dataset from physical samples to predict the curvatures of self-shaping wood bilayers based on local variations in the grain. In contrast to state-of-the-art predictive models, specifically 1.) a 2D Timoshenko model and 2.) a 3D numerical model with a rheological model, our method accounts for natural and unavoidable material variations. In this paper, we only focus on local grain variations as the main driver for curvatures in small-scale material samples. We extracted a feature matrix from grain images of active and passive layers as a Grey Level Co-Occurrence Matrix and used it as the input for our ML models. We also analysed the impact of grain variations on the feature matrix. We trained and tested several tree-based regression models with different features. The models achieved very accurate predictions for curvatures in each sample (R;0.9) and extend the range of parameters that is incalculable by a Timoshenko model. This research contributes to the material-efficient design of weather-responsive shape-changing wood structures by further leveraging the use of natural material features and explainable data-driven modelling and extends the topic in ML for material behaviour-driven design among the CAADRIA community.
keywords	data-driven model, machine learning, material programming, smart material, timber structure, SDG 12
series	CAADRIA
email
last changed	2022/07/22 07:34

_id	cdrf2022_253
id	cdrf2022_253
authors	Chuheng Tan and Ximing Zhong
year	2022
title	A Rapid Wind Velocity Prediction Method in Built Environment Based on CycleGAN Model
doi	https://doi.org/https://doi.org/10.1007/978-981-19-8637-6_22
source	Proceedings of the 2022 DigitalFUTURES The 4st International Conference on Computational Design and Robotic Fabrication (CDRF 2022)
summary	Although the wind microclimate and wind environment play important roles in urban prediction, the time-consuming and complicated setup and process of wind simulation are widely regarded as challenges. There are several methods to use deep learning (DL) models for wind speed prediction by labeling pairs of wind simulation dataset samples. However, many wind simulation experiments are needed to obtain paired datasets, which is still time-consuming and cumbersome. Compared with previous studies, we propose a method to train a DL model without labelling paired data, which is based on Cycle Generative Adversarial Network (cycleGAN). To verify our hypothesis, we evaluate the results and process of the pix2pix model (requires paired datasets) and cycleGAN (does not requires paired datasets), and explore the difference of results between these two DL models and professional CFD software. The result shows that cycleGAN can perform as well as pix2pix in accuracy, indicating that some random city plans image samples and random wind simulation samples can train surrogate models as accurate as labelled DL methods. Although the DL method has similar results to the professional CFD method, the details of the wind flow results still need improvement. This study can help designers and policymakers to make informed decisions to choose Dl methods for real-time wind speed prediction for early-stage design exploration.
series	cdrf
email
last changed	2024/05/29 14:02

_id	ascaad2022_065
id	ascaad2022_065
authors	David, Joao; Leitao, Antonio
year	2022
title	Getting a Handle on Floor Plan Analysis: Door Classification in Floor Plans and a Survey on Existing Datasets
source	Hybrid Spaces of the Metaverse - Architecture in the Age of the Metaverse: Opportunities and Potentials [10th ASCAAD Conference Proceedings] Debbieh (Lebanon) [Virtual Conference] 12-13 October 2022, pp. 221-236
summary	Floor plan interpretation and reconstruction is crucial to enable the transformation of drawings to 3D models or different digital formats. It has recently taken advantage of neural-based architectures, especially in the semantic segmentation field. These techniques perform better than traditional methods, but the results depend mainly on the data used to train the networks, which is often crafted for the specific task being performed, making it hard to reuse for different purposes. In this paper, we conduct a literature survey on the existing datasets for floor plan analysis, and we explore how information regarding door placement and orientation can be recovered without having to change the initial data or model. We propose a two-step recognition method based on image segmentation followed by classification of cropped zones to allow data augmentation during training. In the process, we generate a dataset consisting of 35000 annotated door images extracted from an existing dataset.
series	ASCAAD
email
last changed	2024/02/16 13:29

_id	ecaade2022_175
id	ecaade2022_175
authors	Di Carlo, Raffaele, Mittal, Divyae and Vesely, Ondrej
year	2022
title	Generating 3D Building Volumes for a Given Urban Context using Pix2Pix GAN
doi	https://doi.org/10.52842/conf.ecaade.2022.2.287
source	Pak, B, Wurzer, G and Stouffs, R (eds.), Co-creating the Future: Inclusion in and through Design - Proceedings of the 40th Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2022) - Volume 2, Ghent, 13-16 September 2022, pp. 287–295
summary	Our ability to delegate the most intellectually demanding tasks to machines improves with each passing day. Even in the fields of architecture and design, which were previously thought to be exclusive domain of human creativity and flare, we are moving the first steps towards developing models that can capture the patterns, invisible to the naked eye, embedded in the creative process. These patterns reflect ideas and traditions, imprinted in the collective mind over the course of history, that can be improved upon or serve as a cautionary tale for the new generation of designers in their work of designing an equitable, more inclusive future. Generative Adversarial Networks (GANs) give us the opportunity to turn style and design into learnable features that can be used to automatically generate blueprints and layouts. In this study, we attempt to apply this technology to urban design and to the task of generating a building footprint and volume that fits within the surrounding built environment. We do so by developing a Pix2Pix model composed of a ResNet-6 generator and a Patch discriminator, applying it to satellite views of neighborhoods from across the Netherlands, and then turning the resulting 2D generated building footprint into a reusable 3D model. The model is trained using the national cadastral data and TU Delft 3D BAG dataset. The results show that it is possible to predict a building shape compatible in style and height with the surroundings. Although the model can be used for different applications, we use it as an evaluation tool to compare the design alternatives fitting the desired contextual patterns.
keywords	Generative Adversarial Networks, Urban Design, Pix2Pix, Raster Vectorization, 3D Rendering
series	eCAADe
email
last changed	2024/04/22 07:10

_id	caadria2022_114
id	caadria2022_114
authors	Dong, Zhiyong, Lin, Jinru, Wang, Siqi, Xu, Yijia, Xu, Jiaqi and Liu, Xiao
year	2022
title	Where Will Romance Occur, A New Prediction Method of Urban Love Map through Deep Learning
doi	https://doi.org/10.52842/conf.caadria.2022.1.213
source	Jeroen van Ameijde, Nicole Gardner, Kyung Hoon Hyun, Dan Luo, Urvi Sheth (eds.), POST-CARBON - Proceedings of the 27th CAADRIA Conference, Sydney, 9-15 April 2022, pp. 213-222
summary	Romance awakens fond memories of the city. Finding out the relationship between romantic scene and urban morphology, and providing a prediction, can potentially facilitate the better urban design and urban life. Taking the Yangtze River Delta region of China as an example, this study aims to predict the distribution of romantic locations using deep learning based on multi-source data. Specifically, we use web crawlers to extract romance-related messages and geographic locations from social media platforms, and visualize them as romance heatmap. The urban environment and building features associated with romantic information are identified by Pearson correlation analysis and annotated in the city map. Then, both city labelled maps and romance heatmaps are fed into a Generative Adversarial Networks (GAN) as the training dataset to achieve final romance distribution predictions across regions for other cities. The ideal prediction results highlight the ability of deep learning techniques to quantify experience-based decision-making strategies that can be used in further research on urban design.
keywords	Romance Heatmap, Generative Adversarial Networks, Deep Learning, Big Data Analysis, Correlation Analysis, SDG 11
series	CAADRIA
email
last changed	2022/07/22 07:34

For more results click below: