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 524

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

_id	cf2019_048
id	cf2019_048
authors	Argota Sanchez-Vaquerizo, Javier and Daniel Cardoso Llach
year	2019
title	The Social Life of Small Urban Spaces 2.0 Three Experiments in Computational Urban Studies
source	Ji-Hyun Lee (Eds.) "Hello, Culture!" [18th International Conference, CAAD Futures 2019, Proceedings / ISBN 978-89-89453-05-5] Daejeon, Korea, p. 430
summary	This paper introduces a novel framework for urban analysis that leverages computational techniques, along with established urban research methods, to study how people use urban public space. Through three case studies in different urban locations in Europe and the US, it demonstrates how recent machine learning and computer vision techniques may assist us in producing unprecedently detailed portraits of the relative influence of urban and environmental variables on people’s use of public space. The paper further discusses the potential of this framework to enable empirically-enriched forms of urban and social analysis with applications in urban planning, design, research, and policy.
keywords	Data Analytics, Urban Design, Machine Learning, Artificial Intelligence, Big Data, Space Syntax
series	CAAD Futures
email
last changed	2019/07/29 14:18

_id	cf2019_020
id	cf2019_020
authors	Belém, Catarina; Luís Santos and António Leitão
year	2019
title	On the Impact of Machine Learning: Architecture without Architects?
source	Ji-Hyun Lee (Eds.) "Hello, Culture!" [18th International Conference, CAAD Futures 2019, Proceedings / ISBN 978-89-89453-05-5] Daejeon, Korea, pp. 148-167
summary	Architecture has always followed and adopted technological breakthroughs of other areas. As a case in point, in the last decades, the field of computation changed the face of architectural practice. Considering the recent breakthroughs of Machine Learning (ML), it is expectable to see architecture adopting ML-based approaches. However, it is not yet clear how much this adoption will change the architectural practice and in order to forecast this change it is necessary to understand the foundations of ML and its impact in other fields of human activity. This paper discusses important ML techniques and areas where they were successfully applied. Based on those examples, this paper forecast hypothetical uses of ML in the realm of building design. In particular, we examine ML approaches in conceptualization, algorithmization, modeling, and optimization tasks. In the end, we conjecture potential applications of such approaches, suggest future lines of research, and speculate on the future face of the architectural profession.
keywords	Machine Learning, Algorithmic Design, AI for Building Design
series	CAAD Futures
type	normal paper
email
last changed	2019/07/29 14:54

_id	ijac201917106
id	ijac201917106
authors	Brown, Nathan C. and Caitlin T. Mueller
year	2019
title	Design variable analysis and generation for performance-based parametric modeling in architecture
source	International Journal of Architectural Computing vol. 17 - no. 1, 36-52
summary	Many architectural designers recognize the potential of parametric models as a worthwhile approach to performance- driven design. A variety of performance simulations are now possible within computational design environments, and the framework of design space exploration allows users to generate and navigate various possibilities while considering both qualitative and quantitative feedback. At the same time, it can be difficult to formulate a parametric design space in a way that leads to compelling solutions and does not limit flexibility. This article proposes and tests the extension of machine learning and data analysis techniques to early problem setup in order to interrogate, modify, relate, transform, and automatically generate design variables for architectural investigations. Through analysis of two case studies involving structure and daylight, this article demonstrates initial workflows for determining variable importance, finding overall control sliders that relate directly to performance and automatically generating meaningful variables for specific typologies.
keywords	Parametric design, design space formulation, data analysis, design variables, dimensionality reduction
series	journal
email
last changed	2019/08/07 14:04

_id	cf2019_014
id	cf2019_014
authors	Ferrando, Cecilia; Niccolo Dalmasso, Jiawei Mai, Daniel Cardoso Llach
year	2019
title	Architectural Distant Reading Using Machine Learning to Identify Typological Traits Across Multiple Buildings
source	Ji-Hyun Lee (Eds.) "Hello, Culture!" [18th International Conference, CAAD Futures 2019, Proceedings / ISBN 978-89-89453-05-5] Daejeon, Korea, pp. 114-127
summary	This paper introduces an approach to architectural “distant reading”: the use of computational methods to analyze architectural data in order to derive spatial insights from—and explore new questions concerning—large collections of architectural work. Through a case study comprising a dataset of religious buildings, we show how we may use machine learning techniques to identify typological and functional traits from building plans. We find that spatial structure, rather than local features, is particularly effective in supporting this type of analysis. Further, we speculate on the potential of this computational method to enrich architectural design, research, and criticism by, for example, enabling new ways of thinking about architectural concepts such as typology in ways that reflect gradual variations, rather than sharp distinctions.
keywords	Architectural Analytics, Machine Learning, Classification, Religious buildings, Space Syntax
series	CAAD Futures
email
last changed	2019/07/29 14:08

_id	ecaadesigradi2019_510
id	ecaadesigradi2019_510
authors	Giannopoulou, Effima, Baquero, Pablo, Warang, Angad, Orciuoli, Affonso and T. Estévez, Alberto
year	2019
title	Stripe Segmentation for Branching Shell Structures - A Data Set Development as a Learning Process for Fabrication Efficiency and Structural Performance
doi	https://doi.org/10.52842/conf.ecaade.2019.3.063
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 3, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 63-70
summary	This article explains the evolution towards the subject of digital fabrication of thin shell structures, searching for the computational design techniques which allow to implement biological pattern mechanisms for efficient fabrication procedures. The method produces data sets in order to analyse and evaluate parallel alternatives of branching topologies, segmentation patterns, material usage, weight and deflection values as a user learning process. The importance here is given to the selection of the appropriate attributes, referring to which specific geometric characteristics of the parametric model are affecting each other and with what impact. The outcomes are utilized to train an Artificial Neural Network to predict new building information based on new combinations of desired parameters so that the user can decide and adjust the design based on the new information.
keywords	Digital Fabrication; Shell Structures; Segmentation; Machine Learning; Branching Topologies; Bio-inspired
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:51

_id	acadia19_664
id	acadia19_664
authors	Koshelyuk, Daniil; Talaei, Ardeshir; Garivani, Soroush; Markopoulou, Areti; Chronis, Angelo; Leon, David Andres; Krenmuller, Raimund
year	2019
title	Alive
doi	https://doi.org/10.52842/conf.acadia.2019.664
source	ACADIA 19:UBIQUITY AND AUTONOMY [Proceedings of the 39th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-59179-7] (The University of Texas at Austin School of Architecture, Austin, Texas 21-26 October, 2019) pp. 664-673
summary	In the context of data-driven culture, built space still maintains low responsiveness and adaptability. Part of this reality lies in the low resolution of live information we have about the behavior and condition of surfaces and materials. This research addresses this issue by exploring the development of a deformation-sensing composite membrane material system following a bottom-up approach and combining various technologies toward solving related technical issues—exploring conductivity properties of graphene and maximizing utilization within an architecture-related proof-of-concept scenario and a workflow including design, fabrication, and application methodology. Introduced simulation of intended deformation helps optimize the pattern of graphene nanoplatelets (GNP) to maximize membrane sensitivity to a specific deformation type while minimizing material usage. Research explores various substrate materials and graphene incorporation methods with initial geometric exploration. Finally, research introduces data collection and machine learning techniques to train recognition of certain types of deformation (single point touch) on resistance changes. The final prototype demonstrates stable and symmetric readings of resistance in a static state and, after training, exhibits an 88% prediction accuracy of membrane shape on a labeled sample data-set through a pre-trained neural network. The proposed framework consisting of a simulation based, graphene-capturing fabrication method on stretchable surfaces, and includes initial exploration in neural network training shape detection, which combined, demonstrate an advanced approach to embedding intelligence.
series	ACADIA
type	normal paper
email
last changed	2022/06/07 07:51

_id	ecaade2024_92
id	ecaade2024_92
authors	Mayor Luque, Ricardo; Beguin, Nestor; Rizvi Riaz, Sheikh; Dias, Jessica; Pandey, Sneham
year	2024
title	Multi-material Gradient Additive Manufacturing: A data-driven performative design approach to multi-materiality through robotic fabrication
doi	https://doi.org/10.52842/conf.ecaade.2024.1.381
source	Kontovourkis, O, Phocas, MC and Wurzer, G (eds.), Data-Driven Intelligence - Proceedings of the 42nd Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2024), Nicosia, 11-13 September 2024, Volume 1, pp. 381–390
summary	Buildings are responsible for 39% of global energy-related carbon emissions, with operational activities contributing 28% and materials and construction accounting for 11%(World Green Building Council, 2019) It is therefore vital to reconsider our reliance on fossil fuels for building materials and to develop new advanced manufacturing techniques that enable an integrated approach to material-controlled conception and production. The emergence of Multi-material Additive Manufacturing (MM-AM) technology represents a paradigm shift in producing elements with hybrid properties derived from novel and optimized solutions. Through robotic fabrication, MM-AM offers streamlined operations, reduced material usage, and innovative fabrication methods. It encompasses a plethora of methods to address diverse construction needs and integrates material gradients through data-driven analyses, challenging traditional prefabrication practices and emphasizing the current growth of machine learning algorithms in design processes. The research outlined in this paper presents an innovative approach to MM-AM gradient 3D printing through robotic fabrication, employing data-driven performative analyses enabling control over print paths for sustainable applications in both the AM industry and our built environment. The article highlights several designed prototypes from two distinct phases, demonstrating the framework's viability, implications, and constraints: a workshop dedicated to data-driven analyses in facade systems for MM-AM 3D-printed brick components, and a 3D-printed brick facade system utilizing two renewable and bio-materials—Cork sourced from recycled stoppers and Charcoal, with the potential for carbon sequestration.
keywords	Data-driven Performative design, Multi-material 3d Printing, Material Research, Fabrication-informed Material Design, Robotic Fabrication
series	eCAADe
email
last changed	2024/11/17 22:05

_id	sigradi2023_234
id	sigradi2023_234
authors	Santos, Ítalo, Andrade, Max, Zanchettin, Cleber and Rolim, Adriana
year	2023
title	Machine learning applied in the evaluation of airport projects in Brazil based on BIM models
source	García Amen, F, Goni Fitipaldo, A L and Armagno Gentile, Á (eds.), Accelerated Landscapes - Proceedings of the XXVII International Conference of the Ibero-American Society of Digital Graphics (SIGraDi 2023), Punta del Este, Maldonado, Uruguay, 29 November - 1 December 2023, pp. 875–887
summary	In a country with continental dimensions like Brazil, air transport plays a strategic role in the development of the country. In recent years, initiatives have been promoted to boost the development of air transport, among which the BIM BR strategy stands out, instituted by decree n-9.983 (2019), decree n-10.306 (2020) and more recently, the publication of the airport design manual (SAC, 2021). In this context, this work presents partial results of a doctoral research based on the Design Science Research (DSR) method for the application of Machine Learning (ML) techniques in the Artificial Intelligence (AI) subarea, aiming to support SAC airport project analysts in the phase of project evaluation. Based on a set of training and test data corresponding to airport projects, two ML algorithms were trained. Preliminary results indicate that the use of ML algorithms enables a new scenario to be explored by teams of airport design analysts in Brazil.
keywords	Airports, Artificial intelligence, BIM, Evaluation, Machine learning.
series	SIGraDi
email
last changed	2024/03/08 14:07

_id	acadia19_392
id	acadia19_392
authors	Steinfeld, Kyle
year	2019
title	GAN Loci
doi	https://doi.org/10.52842/conf.acadia.2019.392
source	ACADIA 19:UBIQUITY AND AUTONOMY [Proceedings of the 39th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-59179-7] (The University of Texas at Austin School of Architecture, Austin, Texas 21-26 October, 2019) pp. 392-403
summary	This project applies techniques in machine learning, specifically generative adversarial networks (or GANs), to produce synthetic images intended to capture the predominant visual properties of urban places. We propose that imaging cities in this manner represents the first computational approach to documenting the Genius Loci of a city (Norberg-Schulz, 1980), which is understood to include those forms, textures, colors, and qualities of light that exemplify a particular urban location and that set it apart from similar places. Presented here are methods for the collection of urban image data, for the necessary processing and formatting of this data, and for the training of two known computational statistical models (StyleGAN (Karras et al., 2018) and Pix2Pix (Isola et al., 2016)) that identify visual patterns distinct to a given site and that reproduce these patterns to generate new images. These methods have been applied to image nine distinct urban contexts across six cities in the US and Europe, the results of which are presented here. While the product of this work is not a tool for the design of cities or building forms, but rather a method for the synthetic imaging of existing places, we nevertheless seek to situate the work in terms of computer-assisted design (CAD). In this regard, the project is demonstrative of a new approach to CAD tools. In contrast with existing tools that seek to capture the explicit intention of their user (Aish, Glynn, Sheil 2017), in applying computational statistical methods to the production of images that speak to the implicit qualities that constitute a place, this project demonstrates the unique advantages offered by such methods in capturing and expressing the tacit.
series	ACADIA
type	normal paper
email
last changed	2022/06/07 07:56

_id	caadria2019_638
id	caadria2019_638
authors	Willemse, Elias Jakobus, Tuncer, Bige and Bouffanais, Roland
year	2019
title	Identifying Highly Dense Areas from Raw Location Data
doi	https://doi.org/10.52842/conf.caadria.2019.2.805
source	M. Haeusler, M. A. Schnabel, T. Fukuda (eds.), Intelligent & Informed - Proceedings of the 24th CAADRIA Conference - Volume 2, Victoria University of Wellington, Wellington, New Zealand, 15-18 April 2019, pp. 805-814
summary	In this paper we show how very high-volumes of raw WiFi-based location data of individuals can be used to identify dense activity locations within a neighbourhood. Key to our methods is the inference of the size of the area directly from the data, without having to use additional geographical information. To extract the density information, data-mining and machine learning techniques form activity-based transportation modelling are applied. These techniques are demonstrated on data from a large-scale experiment conducted in Singapore in which tens of thousands of school children carried a multi-sensor device for five consecutive days. By applying the techniques we were able to identify expected high-density areas of school pupils, specifically their school locations, using only the raw data, demonstrating the general applicability of the methods.
keywords	; Machine Learning, Big-data, Location-analysis
series	CAADRIA
email
last changed	2022/06/07 07:56

_id	ecaadesigradi2019_116
id	ecaadesigradi2019_116
authors	Fernando, Shayani
year	2019
title	Collaborative Crafting of Interlocking Structures in Stereotomic Practice
doi	https://doi.org/10.52842/conf.ecaade.2019.2.183
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 2, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 183-190
summary	Situated within the art of cutting solids (stereotomy) and the evolution of machine tools; this research will investigate subtractive fabrication in relation to robotic carving of stone structures. The advancement of the industrial revolutions in the mid to late 19th century saw the rise of new building techniques and materials which were primarily based on structural steel construction. The modern aesthetic of the time further diminished the place of traditional stonework and ornamentation in modern structures within the building arts. This paper will focus on the design and fabrication of three sculptural dry-stone modular prototypes investigating interlocking self-supporting structures in stone. Examining the value of robotic technologies in the design and construction process in relation to collaborative crafting of the hand and machine. Accommodating for material tolerances which are a major factor in this research. Interrogating the value of robotic crafting with material implications and exploring the role of the artisan in machine crafted architectural components.
keywords	Collaborative; Crafting; Interlocking; Structures; Robotic Fabrication; Digital Stone
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:50

_id	acadia19_448
id	acadia19_448
authors	Hahm, Soomeen
year	2019
title	Augmented Craftsmanship
doi	https://doi.org/10.52842/conf.acadia.2019.448
source	ACADIA 19:UBIQUITY AND AUTONOMY [Proceedings of the 39th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-59179-7] (The University of Texas at Austin School of Architecture, Austin, Texas 21-26 October, 2019) pp. 448-457
summary	Over the past decade, we have witnessed rapid advancements on both practical and theoretical levels in regard to automated construction as a consequence of increasing sophistication of digital fabrication technologies such as robotics, 3D printing, etc. However, digital fabrication technology is often very limited when it comes to dealing with delicate and complex crafting processes. Although digital fabrication processes have become widely accessible and utilized across industries in recent times, there are still a number of fabrication techniques—which heavily rely on human labour—due to the complex nature of procedures and delicacy of materials. With this in mind, we need to ask ourselves if full automation is truly an ultimate goal, or if we need to (re)consider the role of humans in the architectural construction chain, as automation becomes more prevalent. We propose rethinking the role which human, machine, and computer have in construction— occupying the territory between purely automated, exclusively robotically-driven fabrication and highly crafted processes requiring human labour. This is to propose an alternative to reducing construction to fully automated assembly of simplified/discretized building parts, by appreciating physical properties of materials and nature of crafting processes. The research proposes a design-to-construction workflow pursued and enabled by augmented humans using AR devices. As a result, proposed workflows are tested on three prototypical inhabitable structure, aiming to be applicable to other projects in the near future, and to bridge the gap between purely automated construction processes on one hand, and craft-based, material-driven but labour-intensive processes on the other.
series	ACADIA
type	normal paper
email
last changed	2022/06/07 07:51

_id	cf2019_015
id	cf2019_015
authors	Ladron de Guevara, Manuel; Luis Ricardo Borunda and Ramesh Krishnamurti
year	2019
title	A Multi-Resolution Design Methodology Based on Discrete Models
source	Ji-Hyun Lee (Eds.) "Hello, Culture!" [18th International Conference, CAAD Futures 2019, Proceedings / ISBN 978-89-89453-05-5] Daejeon, Korea, p. 129
summary	The use of programming languages in design opens up unexplored and previously unworkable territories, mainly, in conventional architectural practice. In the 1990s, languages of continuity, smoothness and seamlessness dominated the architectural inquiry with the CNC milling machine as its manufacturing tool. Today’s computational design and fabrication technology look at languages of synthesis of fragments or particles, with the 3D printer as its fabrication archetype. Fundamental to this idea is the concept of resolution– the amount of information stored at any localized region. Construction of a shape is then based on multiple regions of resolution. This paper explores a novel design methodology that takes this concept of resolutions on discrete elements as a design driver for architectural practice. This research has been tested primarily through additive manufacturing techniques.
keywords	Multi-Resolution Design Methodology; Discrete-Based Computational Design; Resolutions; Additive Manufacturing
series	CAAD Futures
email
last changed	2019/07/29 14:08

_id	acadia19_298
id	acadia19_298
authors	Leach, Neil
year	2019
title	Do Robots Dream of Digital Sleep?
doi	https://doi.org/10.52842/conf.acadia.2019.298
source	ACADIA 19:UBIQUITY AND AUTONOMY [Proceedings of the 39th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-59179-7] (The University of Texas at Austin School of Architecture, Austin, Texas 21-26 October, 2019) pp. 298-309
summary	AI is playing an increasingly important role in everyday life. But can AI actually design? This paper takes its point of departure from Philip K Dick’s novel, Do Androids Dream of Electric Sheep? and refers to Google’s DeepDream software, and other AI techniques such as GANs, Progressive GANs, CANs and StyleGAN, that can generate increasingly convincing images, a process often described as ‘dreaming’. It notes that although generative AI does not possess consciousness, and therefore cannot literally dream, it can still be a powerful design tool that becomes a prosthetic extension to the human imagination. Although the use of GANs and other deep learning AI tools is still in its infancy, we are at the dawn of an exciting – but also potentially terrifying – new era for architectural design. Most importantly, the paper concludes, the development of AI is also helping us to understand human intelligence and 'creativity'.
series	ACADIA
type	normal paper
email
last changed	2022/06/07 07:52

_id	caadria2019_000
id	caadria2019_000
authors	M. Haeusler, M. A. Schnabel, T. Fukuda (eds.)
year	2019
title	CAADRIA 2019: Intelligent & Informed, Volume 1
doi	https://doi.org/10.52842/conf.caadria.2019.1
source	Intelligent & Informed - Proceedings of the 24th CAADRIA Conference - Volume 1, Victoria University of Wellington, Wellington, New Zealand, 15-18 April 2019, 830 p.
summary	The territories of computational design are ever-changing and represent a substantial region that remains uncharted; one with expanding and permeable boundaries that continue to be fully breached. This ocean of opportunity implores researchers to embark on ambitious journeys of exploration. Undulating and temporal, computational design needs research that engages explicitly with the innovative, intelligent and informed exploitation of computational design, and with the array of computational technologies that the discipline may engage with. Human intelligence and creativity deliver the hegemonic direction for the field of computer-mediated architectural design research; an area where the computational component is a core aspect of the investigation. The actors in this are both witness to, and instigators of, the exciting, consequent, well-founded research that continues to deliver new knowledge, insights and information. This, then, explains the specific overarching theme of the conference: ‘Intelligent and Informed’. The scope of this theme is driven by the intention to take in aspects of machine intelligence, and a wide range of potential research that engages with the intelligent exploitation of computer-mediated techniques in Architecture.
series	CAADRIA
last changed	2022/06/07 07:49

_id	caadria2019_001
id	caadria2019_001
authors	M. Haeusler, M. A. Schnabel, T. Fukuda (eds.)
year	2019
title	CAADRIA 2019: Intelligent & Informed, Volume 2
doi	https://doi.org/10.52842/conf.caadria.2019.2
source	Intelligent & Informed - Proceedings of the 24th CAADRIA Conference - Volume 2, Victoria University of Wellington, Wellington, New Zealand, 15-18 April 2019, 830 p.
summary	The territories of computational design are ever-changing and represent a substantial region that remains uncharted; one with expanding and permeable boundaries that continue to be fully breached. This ocean of opportunity implores researchers to embark on ambitious journeys of exploration. Undulating and temporal, computational design needs research that engages explicitly with the innovative, intelligent and informed exploitation of computational design, and with the array of computational technologies that the discipline may engage with. Human intelligence and creativity deliver the hegemonic direction for the field of computer-mediated architectural design research; an area where the computational component is a core aspect of the investigation. The actors in this are both witness to, and instigators of, the exciting, consequent, well-founded research that continues to deliver new knowledge, insights and information. This, then, explains the specific overarching theme of the conference: ‘Intelligent and Informed’. The scope of this theme is driven by the intention to take in aspects of machine intelligence, and a wide range of potential research that engages with the intelligent exploitation of computer-mediated techniques in Architecture.
series	CAADRIA
last changed	2022/06/07 07:49

_id	ecaadesigradi2019_135
id	ecaadesigradi2019_135
authors	Newton, David
year	2019
title	Deep Generative Learning for the Generation and Analysis of Architectural Plans with Small Datasets
doi	https://doi.org/10.52842/conf.ecaade.2019.2.021
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 2, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 21-28
summary	The field of generative architectural design has explored a wide range of approaches in the automation of design production, but these approaches have demonstrated limited artificial intelligence. Generative Adversarial Networks (GANs) are a leading deep generative model that use deep neural networks (DNNs) to learn from a set of training examples in order to create new design instances with a degree of flexibility and fidelity that outperform competing generative approaches. Their application to generative tasks in architecture, however, has been limited. This research contributes new knowledge on the use of GANs for architectural plan generation and analysis in relation to the work of specific architects. Specifically, GANs are trained to synthesize architectural plans from the work of the architect Le Corbusier and are used to provide analytic insight. Experiments demonstrate the efficacy of different augmentation techniques that architects can use when working with small datasets.
keywords	generative design; deep learning; artificial intelligence; generative adversarial networks
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:58

_id	caadria2021_053
id	caadria2021_053
authors	Rhee, Jinmo and Veloso, Pedro
year	2021
title	Generative Design of Urban Fabrics Using Deep Learning
doi	https://doi.org/10.52842/conf.caadria.2021.1.031
source	A. Globa, J. van Ameijde, A. Fingrut, N. Kim, T.T.S. Lo (eds.), PROJECTIONS - Proceedings of the 26th CAADRIA Conference - Volume 1, The Chinese University of Hong Kong and Online, Hong Kong, 29 March - 1 April 2021, pp. 31-40
summary	This paper describes the Urban Structure Synthesizer (USS), a research prototype based on deep learning that generates diagrams of morphologically consistent urban fabrics from context-rich urban datasets. This work is part of a larger research on computational analysis of the relationship between urban context and morphology. USS relies on a data collection method that extracts GIS data and converts it to diagrams with context information (Rhee et al., 2019). The resulting dataset with context-rich diagrams is used to train a Wasserstein GAN (WGAN) model, which learns how to synthesize novel urban fabric diagrams with the morphological and contextual qualities present in the dataset. The model is also trained with a random vector in the input, which is later used to enable parametric control and variation for the urban fabric diagram. Finally, the resulting diagrams are translated to 3D geometric entities using computer vision techniques and geometric modeling. The diagrams generated by USS suggest that a learning-based method can be an alternative to methods that rely on experts to build rule sets or parametric models to grasp the morphological qualities of the urban fabric.
keywords	Deep Learning; Urban Fabric; Generative Design; Artificial Intelligence; Urban Morphology
series	CAADRIA
email
last changed	2022/06/07 07:56

_id	caadria2020_259
id	caadria2020_259
authors	Rhee, Jinmo, Veloso, Pedro and Krishnamurti, Ramesh
year	2020
title	Integrating building footprint prediction and building massing - an experiment in Pittsburgh
doi	https://doi.org/10.52842/conf.caadria.2020.2.669
source	D. Holzer, W. Nakapan, A. Globa, I. Koh (eds.), RE: Anthropocene, Design in the Age of Humans - Proceedings of the 25th CAADRIA Conference - Volume 2, Chulalongkorn University, Bangkok, Thailand, 5-6 August 2020, pp. 669-678
summary	We present a novel method for generating building geometry using deep learning techniques based on contextual geometry in urban context and explore its potential to support building massing. For contextual geometry, we opted to investigate the building footprint, a main interface between urban and architectural forms. For training, we collected GIS data of building footprints and geometries of parcels from Pittsburgh and created a large dataset of Diagrammatic Image Dataset (DID). We employed a modified version of a VGG neural network to model the relationship between (c) a diagrammatic image of a building parcel and context without the footprint, and (q) a quadrilateral representing the original footprint. The option for simple geometrical output enables direct integration with custom design workflows because it obviates image processing and increases training speed. After training the neural network with a curated dataset, we explore a generative workflow for building massing that integrates contextual and programmatic data. As trained model can suggest a contextual boundary for a new site, we used Massigner (Rhee and Chung 2019) to recommend massing alternatives based on the subtraction of voids inside the contextual boundary that satisfy design constraints and programmatic requirements. This new method suggests the potential that learning-based method can be an alternative of rule-based design methods to grasp the complex relationships between design elements.
keywords	Deep Learning; Prediction; Building Footprint; Massing; Generative Design
series	CAADRIA
email
last changed	2022/06/07 07:56

_id	ecaadesigradi2019_630
id	ecaadesigradi2019_630
authors	Saad, Carla
year	2019
title	If Only Wood Could Speak... - Explorations in digital fabrication processes based on timber grain and patterns
doi	https://doi.org/10.52842/conf.ecaade.2019.3.227
source	Sousa, JP, Xavier, JP and Castro Henriques, G (eds.), Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference - Volume 3, University of Porto, Porto, Portugal, 11-13 September 2019, pp. 227-234
summary	This paper presents an exploration following wood patterns applied to traditional woodworking techniques such as steam bending and carving. It builds on a digital fabrication process where the patterns that are unique to each wood strip and which constitutes their structural geometry are also used within fabrication processes and applied machinery such as the zund machine. The experiments were done in two series where the earlier set focuses on routing along the wood patterns of thin wood strips and steam bending the different pieces. The second set experimented with carving thick wood strips on passes generated from the extracted patterns. An automated technique based in computer vision and specifically OpenCV was developed using different filters in order to extract the patterns and transfer the results into CAD software where the lines where further manipulated for variable exploration. The results of this study led to a better understanding of wood patterns from a geometrical perspective thus enriching the aesthetic composition of a design while being authentic to the unique nature of each piece of wood.
keywords	Timber; pattern recognition; grain; image; carving ; digital fabrication
series	eCAADeSIGraDi
email
last changed	2022/06/07 07:56

For more results click below: