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 787

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

_id	ecaade2023_392
id	ecaade2023_392
authors	Johanes, Mikhael and Huang, Jeffrey
year	2023
title	Generative Isovist Transformer: Machine learning for spatial sequence synthesis
doi	https://doi.org/10.52842/conf.ecaade.2023.2.471
source	Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 2, Graz, 20-22 September 2023, pp. 471–480
summary	While isovists have been used widely to quantify and analyze architectural space, its utilization for generative design still needs to be explored. On the other hand, advanced deep learning has shown opportunities for data-driven generative design. This research revisits the isovist capacity to represent architecture as a series of spatial sequences and extends the role of isovists beyond merely a perception model to projective agents. This paper presents the development of GIsT: Generative Isovists Transformer in sampling, learning, and generating architectural spatial sequences. By coupling isovists with discrete representation and generative deep learning models, we untapped the generative potential of isovist representation for spatial sequence synthesis. We demonstrated its capacity to learn the architectural spatial sequence and extendability via few-shots learning. The results show a promising direction toward integrating data-driven experiential spatial synthesis in future computational design tools.
keywords	Isovist, Spatial sequence, Generative Design, Discrete representation learning, Transformers, Machine Learning
series	eCAADe
email
last changed	2023/12/10 10:49

_id	ijac202321205
id	ijac202321205
authors	Zhuang, Xinwei; Ju, Yi; Yang, Allen; Caldas, Luisa
year	2023
title	Synthesis and generation for 3D architecture volume with generative modeling
source	International Journal of Architectural Computing 2023, Vol. 21 - no. 2, 297–314
summary	Generative design in architecture has long been studied, yet most algorithms are parameter-based and require explicit rules, and the design solutions are heavily experience-based. In the absence of a real understanding of the generation process of designing architecture and consensus evaluation matrices, empirical knowledge may be difficult to apply to similar projects or deliver to the next generation. We propose a workflow in the early design phase to synthesize and generate building morphology with artificial neural networks. Using 3D building models from the financial district of New York City as a case study, this research shows that neural networks can capture the implicit features and styles of the input dataset and create a population of design solutions that are coherent with the styles. We constructed our database using two different data representation formats, voxel matrix and signed distance function, to investigate the effect of shape representations on the performance of the generation of building shapes. A generative adversarial neural network and an auto decoder were used to generate the volume. Our study establishes the use of implicit learning to inform the design solution. Results show that both networks can grasp the implicit building forms and generate them with a similar style to the input data, between which the auto decoder with signed distance function representation provides the highest resolution results.
keywords	data-driven design, 3D deep learning, architecture morphology representation, auto decoder, generative adversarial neural network
series	journal
last changed	2024/04/17 14:30

_id	ecaade2023_71
id	ecaade2023_71
authors	Austern, Guy, Yosifof, Roei and Fisher-Gewirtzman, Dafna
year	2023
title	A Dataset for Training Machine Learning Models to Analyze Urban Visual Spatial Experience
doi	https://doi.org/10.52842/conf.ecaade.2023.2.781
source	Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 2, Graz, 20-22 September 2023, pp. 781–790
summary	Previous studies have described the effects of urban attributes such as the Spatial Openness Index (SOI) on pedestrians’ experience. SOI uses 3-dimensional ray casting to quantify the volume of visible space from a single viewpoint. The higher the SOI value, the higher the perceived openness and the lower the perceived density. However, the ray casting simulation on an urban-sized sampling grid is computationally intensive, making this method difficult to use in real-time design tools. Convolutional Neural Networks (CNN), have excellent performance in computer vision in image processing applications. They can be trained to predict the SOI analysis for large urban fabrics in real-time. However, these supervised learning models need a substantial amount of labeled data to train on. For this purpose, we developed a method to generate a large series of height maps and SOI maps of urban fabrics in New York City and encoded them as images using colour information. These height map - SOI analysis image pairs can be used as training data for a CNN to provide rapid, precise visibility simulations on an urban scale.
keywords	Visibility Analysis, Machine Learning, CNN, Perceived Density
series	eCAADe
email
last changed	2023/12/10 10:49

_id	ascaad2023_075
id	ascaad2023_075
authors	Aljhadali, Abdulrahman; Megahed, Yasser; Gwilliam, Julie
year	2023
title	Artificial Intelligence (AI) and Machine Learning (ML) in Practice: A Comprehensive Investigation into the Utilization of Generative Artificial Intelligence (AI) and Machine Learning (ML) in Architectural Practice
source	C+++: Computation, Culture, and Context – Proceedings of the 11th International Conference of the Arab Society for Computation in Architecture, Art and Design (ASCAAD), University of Petra, Amman, Jordan [Hybrid Conference] 7-9 November 2023, pp. 324-343.
summary	This study offers a comprehensive investigation into the utilization of artificial intelligence (AI) and machine learning (ML) technologies within architectural practices. Since the introduction of computer-aided design (CAD), technology has had a significant impact on the way architects conduct their work. This study explores the potential of AI/ML in actual architectural workflows, with a particular emphasis on the capacity of deep neural networks to assist in the design process.The outcome will help to develop a clearer picture of the opportunities and barriers associated with AI for architects; they will also inform the prioritization of focus for future development of this technology in architectural practice, as well as identifying the specific tasks and project phases in which ML could play a role. This research reviewed literature to explore various approaches for applying AI/ML technologies within the field of architecture. Also , complemented by a number of interviews to investigate the ways in which participants are currently using AI/ML in their work, framing the current feedback and the future potential of AI/ML technologies in architecture. The data collection methods adopted involved semi-structured one-on-one interviews with professionals from multi-regional architecture firms and AI developers. The architects interviewed exhibited diverse ways of benefiting from AI/ML technology, with varying approaches and some common trends. The findings demonstrate that AI has played a pivotal role in expediting the design process and enhancing visualization within the field. However, it has also raised concerns, particularly in the realm of privacy.
series	ASCAAD
email
last changed	2024/02/13 14:40

_id	ecaade2023_125
id	ecaade2023_125
authors	Baºarir, Lale, Çiçek, Selen and Koç, Mustafa
year	2023
title	Demystifying the patterns of local knowledge: The implicit relation of local music and vernacular architecture
doi	https://doi.org/10.52842/conf.ecaade.2023.2.791
source	Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 1, Graz, 20-22 September 2023, pp. 791–800
summary	As the zeitgeist suggests, the development of novel design output using Artificial Neural Networks (ANNs) is becoming an important milestone in the architectural design discourse. With the recent encounter of the computational design realm with the diffusion models, it becomes even easier to generate 2D and 3D design outputs. Yet, the utilization of machine learning tools within design computing domains is confined to generating or classifying visual and encoded data. However, it is critical to evaluate the untapped potentials of machine learning technologies in terms of illuminating the implicit correlations and links underlying distinct concepts and themes across a wide range of technical domains. With the ongoing research project named “Local Intelligence", we hypothesized that the local knowledge of a certain location might be conceptualized as a distributed network to connect different forms of local knowledge. As the first case of the project, we tried to reinstate a commonality between the local music and vernacular architecture, for which we trained generative adversarial network (GAN) models with the visual spectrograms translated from the audio data of the local songs and images of vernacular architectural instances from a defined geography. The two multi-modal GAN models differ in terms of the inherent convolutional layers and data pairing process. The outcomes demonstrated that both GAN models can learn how to depict vernacular architectural features from the rhythmic pattern of the songs in various patterns. Consequently, the implicit relations between music and architecture in the initial findings come one step closer to being demystified. Thus, the process and generative outcomes of the two models are compared and discussed in terms of the legibility of the architectural features, by taking the original vernacular architectural image dataset as the ground truth.
keywords	Local Intelligence, Machine Learning, Generative Adversarial Network (GAN), Local Music, Vernacular Architecture
series	eCAADe
email
last changed	2023/12/10 10:49

_id	ecaade2023_000
id	ecaade2023_000
authors	Dokonal, Wolfgang, Hirschberg, Urs and Wurzer, Gabriel
year	2023
title	eCAADe 2023 Digital Design Reconsidered - Volume 1
doi	https://doi.org/10.52842/conf.ecaade.2023.1.001
source	Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 1, Graz, 20-22 September 2023, 905 p.
summary	The conference logo is a bird’s eye view of spiral stairs that join and separate – an homage to the famous double spiral staircase in Graz, a tourist attraction of this city and a must-see for any architecturally minded visitor. Carved out of limestone, the medieval construction of the original is a daring feat of masonry as well as a symbolic gesture. The design speaks of separation and reconciliation: The paths of two people that climb the double spiral stairs separate and then meet again at each platform. The relationship between architectural design and the growing digital repertoire of tools and possibilities seems to undergo similar cycles of attraction and rejection: enthusiasm about digital innovations – whether in Virtual Reality, Augmented Reality, Energy Design, Robotic Fabrication, the many Dimensions of BIM or, as right now, in AI and Machine Learning – is typically followed by a certain disillusionment and a realization that the promises were somewhat overblown. But a turn away from these digital innovations can only be temporary. In our call for papers we refer to the first and second ‘digital turns’, a term Mario Carpo coined. Yes, it’s a bit of a pun, but you could indeed see these digital turns in our logo as well. Carpo would probably agree that design and the digital have become inseparably intertwined. While they may be circling in different directions, an innovative rejoinder is always just around the corner. The theme of the conference asked participants to re-consider the relationship between Design and the Digital. The notion of a cycle is already present in the syllable “re”. Indeed, 20 years earlier, in 2003, we held an ECAADE conference in Graz simply under the title “Digital Design” and our re-using – or is it re-cycling? – the theme can be seen as the completion of one of those cycles described above: One level up, we meet again, we’ve come full circle. The question of the relationship between Design and the Digital is still in flux, still worthy of renewed consideration. There is a historical notion implicit in the theme. To reconsider something, one needs to take a step back, to look into the past as well as into the future. Indeed, at this conference we wanted to take a longer view, something not done often enough in the fast-paced world of digital technology. Carefully considering one’s past can be a source of inspiration. In fact, the double spiral stair that inspired our conference logo also inspired many architects through the ages. Konrad Wachsmann, for example, is said to have come up with his famous Grapevine assembly system based on this double spiral stair and its intricate joinery. More recently, Rem Koolhaas deemed the double spiral staircase in Graz important enough to include a detailed model of it in his “elements of architecture” exhibition at the Venice Biennale in 2014. Our interpretation of the stair is a typically digital one, you might say. First of all: it’s a rendering of a virtual model; it only exists inside a computer. Secondly, this virtual model isn’t true to the original. Instead, it does what the digital has made so easy to do: it exaggerates. Where the original has just two spiral stairs that separate and join, our model consists of countless stairs that are joined in this way. We see only a part of the model, but the stairs appear to continue in all directions. The implication is of an endless field of spiral stairs. As the 3D model was generated with a parametric script, it would be very easy to change all parameters of it – including the number of stairs that make it up. Everyone at this conference is familiar with the concept of parametric design: it makes generating models of seemingly endless amounts of connected spiral stairs really easy. Although, of course, if we’re too literal about the term ‘endless’, generating our stair model will eventually crash even the most advanced computers. We know that, too. – That's another truth about the Digital: it makes a promise of infinity, which, in the end, it can’t keep. And even if it could: what’s the point of just adding more of the same: more variations, more options, more possible ways to get lost? Doesn’t the original double spiral staircase contain all those derivatives already? Don’t we know that ‘more’ isn’t necessarily better? In the original double spiral stair the happy end is guaranteed: the lovers’ paths meet at the top as well as when they exit the building. Therefore, the stair is also colloquially known as the Busserlstiege (the kissing stair) or the Versöhnungsstiege (reconciliation stair). In our digitally enhanced version, this outcome is no longer clear: we can choose between multiple directions at each level and we risk losing sight of the one we were with. This is also emblematic of our field of research. eCAADe was founded to promote “good practice and sharing information in relation to the use of computers in research and education in architecture and related professions” (see ecaade.org). That may have seemed a straightforward proposition forty years ago, when the association was founded. A look at the breadth and depth of research topics presented and discussed at this conference (and as a consequence in this book, for which you’re reading the editorial) shows how the field has developed over these forty years. There are sessions on Digital Design Education, on Digital Fabrication, on Virtual Reality, on Virtual Heritage, on Generative Design and Machine Learning, on Digital Cities, on Simulation and Digital Twins, on BIM, on Sustainability, on Circular Design, on Design Theory and on Digital Design Experimentations. We hope you will find what you’re looking for in this book and at the conference – and maybe even more than that: surprising turns and happy encounters between Design and the Digital.
series	eCAADe
email
last changed	2023/12/10 10:49

_id	ecaade2023_001
id	ecaade2023_001
authors	Dokonal, Wolfgang, Hirschberg, Urs and Wurzer, Gabriel
year	2023
title	eCAADe 2023 Digital Design Reconsidered - Volume 2
doi	https://doi.org/10.52842/conf.ecaade.2023.2.001
source	Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 2, Graz, 20-22 September 2023, 899 p.
summary	The conference logo is a bird’s eye view of spiral stairs that join and separate – an homage to the famous double spiral staircase in Graz, a tourist attraction of this city and a must-see for any architecturally minded visitor. Carved out of limestone, the medieval construction of the original is a daring feat of masonry as well as a symbolic gesture. The design speaks of separation and reconciliation: The paths of two people that climb the double spiral stairs separate and then meet again at each platform. The relationship between architectural design and the growing digital repertoire of tools and possibilities seems to undergo similar cycles of attraction and rejection: enthusiasm about digital innovations – whether in Virtual Reality, Augmented Reality, Energy Design, Robotic Fabrication, the many Dimensions of BIM or, as right now, in AI and Machine Learning – is typically followed by a certain disillusionment and a realization that the promises were somewhat overblown. But a turn away from these digital innovations can only be temporary. In our call for papers we refer to the first and second ‘digital turns’, a term Mario Carpo coined. Yes, it’s a bit of a pun, but you could indeed see these digital turns in our logo as well. Carpo would probably agree that design and the digital have become inseparably intertwined. While they may be circling in different directions, an innovative rejoinder is always just around the corner. The theme of the conference asked participants to re-consider the relationship between Design and the Digital. The notion of a cycle is already present in the syllable “re”. Indeed, 20 years earlier, in 2003, we held an ECAADE conference in Graz simply under the title “Digital Design” and our re-using – or is it re-cycling? – the theme can be seen as the completion of one of those cycles described above: One level up, we meet again, we’ve come full circle. The question of the relationship between Design and the Digital is still in flux, still worthy of renewed consideration. There is a historical notion implicit in the theme. To reconsider something, one needs to take a step back, to look into the past as well as into the future. Indeed, at this conference we wanted to take a longer view, something not done often enough in the fast-paced world of digital technology. Carefully considering one’s past can be a source of inspiration. In fact, the double spiral stair that inspired our conference logo also inspired many architects through the ages. Konrad Wachsmann, for example, is said to have come up with his famous Grapevine assembly system based on this double spiral stair and its intricate joinery. More recently, Rem Koolhaas deemed the double spiral staircase in Graz important enough to include a detailed model of it in his “elements of architecture” exhibition at the Venice Biennale in 2014. Our interpretation of the stair is a typically digital one, you might say. First of all: it’s a rendering of a virtual model; it only exists inside a computer. Secondly, this virtual model isn’t true to the original. Instead, it does what the digital has made so easy to do: it exaggerates. Where the original has just two spiral stairs that separate and join, our model consists of countless stairs that are joined in this way. We see only a part of the model, but the stairs appear to continue in all directions. The implication is of an endless field of spiral stairs. As the 3D model was generated with a parametric script, it would be very easy to change all parameters of it – including the number of stairs that make it up. Everyone at this conference is familiar with the concept of parametric design: it makes generating models of seemingly endless amounts of connected spiral stairs really easy. Although, of course, if we’re too literal about the term ‘endless’, generating our stair model will eventually crash even the most advanced computers. We know that, too. – That's another truth about the Digital: it makes a promise of infinity, which, in the end, it can’t keep. And even if it could: what’s the point of just adding more of the same: more variations, more options, more possible ways to get lost? Doesn’t the original double spiral staircase contain all those derivatives already? Don’t we know that ‘more’ isn’t necessarily better? In the original double spiral stair the happy end is guaranteed: the lovers’ paths meet at the top as well as when they exit the building. Therefore, the stair is also colloquially known as the Busserlstiege (the kissing stair) or the Versöhnungsstiege (reconciliation stair). In our digitally enhanced version, this outcome is no longer clear: we can choose between multiple directions at each level and we risk losing sight of the one we were with. This is also emblematic of our field of research. eCAADe was founded to promote “good practice and sharing information in relation to the use of computers in research and education in architecture and related professions” (see ecaade.org). That may have seemed a straightforward proposition forty years ago, when the association was founded. A look at the breadth and depth of research topics presented and discussed at this conference (and as a consequence in this book, for which you’re reading the editorial) shows how the field has developed over these forty years. There are sessions on Digital Design Education, on Digital Fabrication, on Virtual Reality, on Virtual Heritage, on Generative Design and Machine Learning, on Digital Cities, on Simulation and Digital Twins, on BIM, on Sustainability, on Circular Design, on Design Theory and on Digital Design Experimentations. We hope you will find what you’re looking for in this book and at the conference – and maybe even more than that: surprising turns and happy encounters between Design and the Digital.
series	eCAADe
type	normal paper
email
last changed	2024/08/29 08:36

_id	ecaade2023_145
id	ecaade2023_145
authors	Dortheimer, Jonathan, Schubert, Gerhard, Dalach, Agata, Brenner, Lielle Joy and Martelaro, Nikolas
year	2023
title	Think AI-side the Box! Exploring the Usability of text-to-image generators for architecture students
doi	https://doi.org/10.52842/conf.ecaade.2023.2.567
source	Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 2, Graz, 20-22 September 2023, pp. 567–576
summary	This study examines how architecture students use generative AI image generating models for architectural design. A workshop was conducted with 25 participants to create designs using three state-of-the-art generative diffusion models and BIM or 3D modeling software. Results showed that the participants found the image-generating models useful for the preliminary design stages but had difficulty when the design advanced because the models did not perform as they expected. Finally, the study shows areas for improvement that merit further research. The paper provides empirical evidence on how generative diffusion models are used in an architectural context and contributes to the field of digital design.
keywords	Machine Learning, Diffusion Models, Design Process, Computational Creativity
series	eCAADe
email
last changed	2023/12/10 10:49

_id	caadria2023_446
id	caadria2023_446
authors	Guida, George
year	2023
title	Multimodal Architecture: Applications of Language in a Machine Learning Aided Design Process
doi	https://doi.org/10.52842/conf.caadria.2023.2.561
source	Immanuel Koh, Dagmar Reinhardt, Mohammed Makki, Mona Khakhar, Nic Bao (eds.), HUMAN-CENTRIC - Proceedings of the 28th CAADRIA Conference, Ahmedabad, 18-24 March 2023, pp. 561–570
summary	Recent advances in Natural Language Processing (NLP) and Diffusion Models (DMs) are leading to a significant change in the way architecture is conceived. With capabilities that surpass those of current generative models, it is now possible to produce an unlimited number of high-quality images (Dhariwal and Nichol 2021). This opens up new opportunities for using synthetic images and marks a new phase in the creation of multimodal 3D forms, central to architectural concept design stages. Presented here are three methodologies of generation of meaningful 2D and 3D designs, merging text-to-image diffusion models Stable Diffusion, and DALL-E 2 with computational methods. These allow designers to intuitively navigate through a multimodal feedback loop of information originating from language and aided by artificial intelligence tools. This paper contributes to our understanding of machine-augmented design processes and the importance of intuitive user interfaces (UI) in enabling new dialogues between humans and machines. Through the creation of a prototype of an accessible UI, this exchange of information can empower designers, build trust in these tools, and increase control over the design process.
keywords	Machine Learning, Diffusion Models, Concept Design, Semantics, User Interface, Design Agency
series	CAADRIA
email
last changed	2023/06/15 23:14

_id	acadia23_v3_179
id	acadia23_v3_179
authors	Jabi, Wassim; Leon, David Andres; Alymani, Abdulrahman; Behzad, Selda Pourali; Salamoun, Michelle
year	2023
title	Exploring Building Topology Through Graph Machine Learning
source	ACADIA 2023: Habits of the Anthropocene: Scarcity and Abundance in a Post-Material Economy [Volume 3: Proceedings of the 43rd Annual Conference for the Association for Computer Aided Design in Architecture (ACADIA) ISBN 979-8-9891764-1-0]. Denver. 26-28 October 2023. edited by A. Crawford, N. Diniz, R. Beckett, J. Vanucchi, M. Swackhamer 24-32.
summary	Graph theory offers a powerful method for analyzing complex networks and relationships. When combined with machine learning, graph theory can provide valuable insights into the data generated by 3D models. This workshop integrated advanced spatial modeling and analysis with artificial intelligence, highlighting the importance of technological advancements in shaping the future of architecture and design. It introduced participants to novel workflows that link parametric 3D modeling with concepts of topology, graph theory, and graph machine learning. We used Topologicpy, an advanced spatial modeling and analysis software library designed for Architecture, Engineering, and Construction, paired with DGL, a powerful machine learning library that provides tools for implementing and optimizing graph neural networks (Figure 1). In essence, this process blends cutting-edge technologies and architectural principles that will shape the future of design. Participants learned how to use these workflows to convert 3D models into graphs, analyze their properties, and perform classification and regression tasks. Participants also explored how to create synthetic datasets based on generative and parametric workflows, and build and optimize graph neural networks for specific tasks.
series	ACADIA
type	workshop
last changed	2024/04/17 14:00

_id	ecaade2023_205
id	ecaade2023_205
authors	Meeran, Ahmed and Joyce, Sam
year	2023
title	Rethinking Airport Spatial Analysis and Design: A GAN based data driven approach using latent space exploration on aerial imagery for adaptive airport planning
doi	https://doi.org/10.52842/conf.ecaade.2023.2.501
source	Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 2, Graz, 20-22 September 2023, pp. 501–510
summary	Airports require long term planning, balancing estimations of future demand against available airfield land and site constraints. This is becoming more critical with climate change and the transition to sustainable aviation fuelling infrastructure. This paper demonstrates a novel procedure using Satellite Imagery and Generative Learning to aid in the comparative analysis and early-stage airfield design. Our workflow uses a GAN trained on 2000 images of airports transforming them into a high-dimensional latent space capturing the typologies’ large-scale features. Using a process of projection and dimensional-reduction methods we can locate real-world airport images in the generative latent space and vice-versa. With this capability we can perform comparative “neighbour” analysis at scale based on spatial similarity of features like airfield configuration, and surrounding context. Using this low-dimensional 3D ‘airport designs space’ with meaningful markers provided by existing airports allows for ‘what if’ modelling, such as visualizing an airport on a site without one, modifying an existing airport towards another target airport, or exploring changes in terrain, such as due to climate change or urban development. We present this method a new way to undertake case study, site identification and analysis, as well as undertake speculative design powered by typology informed ML generation, which can be applied to any typologies which could use aerial images to categorize them.
keywords	Airport Development, Machine Learning, GAN, High Dimensional Analysis, Parametric Space Exploration, tSNE, Latent Space Exploration, Data Driven Planning
series	eCAADe
email
last changed	2023/12/10 10:49

_id	ijac202321206
id	ijac202321206
authors	Pouliou, Panagiota; Horvath, Anca-Simona; Palamas, George
year	2023
title	Speculative hybrids: Investigating the generation of conceptual architectural forms through the use of 3D generative adversarial networks
source	International Journal of Architectural Computing 2023, Vol. 21 - no. 2, 315–336
summary	Abstract The process of architectural design aims at solving complex problems that have loosely defined formulations, no explicit basis for terminating the problem-solving activity, and where no ideal solution can be achieved. This means that design problems, as wicked problems, sit in a space between incompleteness and precision. Applying digital tools in general and artificial intelligence in particular to design problems will then mediate solution spaces between incompleteness and precision. In this paper, we present a study where we employed machine learning algorithms to generate conceptual architectural forms for site-specific regulations. We created an annotated dataset of single-family homes and used it to train a 3D Generative Adversarial Network that generated annotated point clouds complying with site constraints. Then, we presented the framework to 23 practitioners of architecture in an attempt to understand whether this framework could be a useful tool for early-stage design. We make a three-fold contribution: First, we share an annotated dataset of architecturally relevant 3D point clouds of single-family homes. Next, we present and share the code for a framework and the results from training the 3D generative neural network. Finally, we discuss machine learning and creative work, including how practitioners feel about the emergence of these tools as mediators between incompleteness and precision in architectural design
keywords	computational design, architecture, machine learning, design process, GNN, point cloud, generative design, artificial intelligence
series	journal
last changed	2024/04/17 14:30

_id	architectural_intelligence2023_5
id	architectural_intelligence2023_5
authors	Qiaoming Deng, Xiaofeng Li, Yubo Liu & Kai Hu
year	2023
title	Exploration of three-dimensional spatial learning approach based on machine learning–taking Taihu stone as an example
doi	https://doi.org/https://doi.org/10.1007/s44223-023-00023-2
source	Architectural Intelligence Journal
summary	Under the influence of globalization, the transformation of traditional architectural space is vital to the growth of local architecture. As an important spatial element of traditional gardens, Taihu stone has the image qualities of being “thin, wrinkled, leaky and transparent” The “transparency” and “ leaky” of Taihu stone reflect the connectivity and irregularity of Taihu stone’s holes, which are consistent with the contemporary architectural design concepts of fluid space and transparency. Nonetheless, relatively few theoretical studies have been conducted on the spatial analysis and design transformation of Taihu stone. Using machine learning, we attempt to extract the three-dimensional spatial variation pattern of Taihu stone in this paper. This study extracts 3D spatial features for experiments using artificial neural networks (ANN) and generative adversarial networks (GAN). In order to extract 3D spatial variation patterns, the machine learning model learns the variation patterns between adjacent sections. The trained machine learning model is capable of generating a series of spatial sections with the spatial variation pattern of the Taihu stone. The purpose of the experimental results is to compare the performance of various machine learning models for 3D space learning in order to identify a model with superior performance. This paper also presents a novel concept for machine learning to master continuous 3D spatial features.
series	Architectural Intelligence
email
last changed	2025/01/09 15:00

_id	ascaad2023_134
id	ascaad2023_134
authors	Salman, Huda; Dounas, Theodoros; Clarke, Connor
year	2023
title	Fluency of Creative Ideas in the Digital Age: Exploring Emergent AI Influences on Design Methodology and Visual Thinking in Architectural Education
source	C+++: Computation, Culture, and Context – Proceedings of the 11th International Conference of the Arab Society for Computation in Architecture, Art and Design (ASCAAD), University of Petra, Amman, Jordan [Hybrid Conference] 7-9 November 2023, pp. 815-832.
summary	Research has explored the concept of originality in visual thinking and architectural education, using different methods. The new state of Artificial Intelligence (AI) in architectural design represents another shift from traditional modes of architectural design and education, into a more authentic approach to the digital age. An experiment is designed to highlight the originality of this approach in design thinking and its futuristic trends and impact on education and creativity studies. The intent of the study we present here is twofold: one to revisit key design studies of design exploration and secondly to explore students' design activity while interacting with text-to-image diffusion machine learning (ML) generative models such as Midjourney, DALL-E and Stable Diffusion, as these might have the potential to change the way that architectural students approach the concept stages of designing projects and products. In addition, we are interested in how the new shift in interfaces and modes of stimulus will influence the students' design process and perceptions. Participants in the design process are final year students who had spent at least four years in a school of architecture and can be classified as semi-experienced designers. Further within the evaluation also lies a critique of the diffusion ML tools themselves as producers of architectonic images, rather than complete concepts for architecture that encapsulate spatial, formal, structural arrangements of elements.
series	ASCAAD
email
last changed	2024/02/13 14:41

_id	ijac202321207
id	ijac202321207
authors	Sebestyen, Adam; Hirschberg, Urs; Rasoulzadeh, Shervin
year	2023
title	Using deep learning to generate design spaces for architecture
source	International Journal of Architectural Computing 2023, Vol. 21 - no. 2, 337–357
summary	We present an early prototype of a design system that uses Deep Learning methodology—Conditional Variational Autoencoders (CVAE)—to arrive at custom design spaces that can be interactively explored using semantic labels. Our work is closely tied to principles of parametric design. We use parametric models to create the dataset needed to train the neural network, thus tackling the problem of lacking 3D datasets needed for deep learning. We propose that the CVAE functions as a parametric tool in itself: The solution space is larger and more diverse than the combined solution spaces of all parametric models used for training. We showcase multiple methods on how this solution space can be navigated and explored, supporting explorations such as object morphing, object addition, and rudimentary 3D style transfer. As a test case, we implemented some examples of the geometric taxonomy of “Operative Design” by Di Mari and Yoo.
keywords	deep learning, generative methods, parametric design, design space, 3D object generation, variational autoencoder, operative design, artificial intelligence, machine learning, voxels
series	journal
last changed	2024/04/17 14:30

_id	caadria2023_24
id	caadria2023_24
authors	Yin, Xiang
year	2023
title	AI and Typology
doi	https://doi.org/10.52842/conf.caadria.2023.1.039
source	Immanuel Koh, Dagmar Reinhardt, Mohammed Makki, Mona Khakhar, Nic Bao (eds.), HUMAN-CENTRIC - Proceedings of the 28th CAADRIA Conference, Ahmedabad, 18-24 March 2023, pp. 39–48
summary	The paper discusses a novel design approach that applies artificial intelligence as an auxiliary tool throughout typology research and architectural design. The method attempts to utilize neural network as a research tool to detect and identify features of a typical architectural type within the specific society context and demonstrate its potential for regional design under the theme of human centric. Typology classification, computational vision, and human-machine collaboration are entwined throughout machine learning and architectural design. The paper aims to demonstrate the ability of 3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions (TreeGAN) to study the inherent principle and characteristic of an architectural type and its potential to provide possible design inspirations based on the typological formation principles concluded by Deep Learning. The article exhibits the key result generated by TreeGAN in a specific architecture type—churches, as the prototype of a design method and conducts a project in Manhattan.
keywords	Architecture Typology, Artificial Intelligence, Machine Learning, TreeGAN, Human-machine Collaboration
series	CAADRIA
email
last changed	2023/06/15 23:14

_id	ecaade2023_166
id	ecaade2023_166
authors	Zhong, Ximing, Koh, Immanuel and Fricker, Pia
year	2023
title	Building-GNN: Exploring a co-design framework for generating controllable 3D building prototypes by graph and recurrent neural networks
doi	https://doi.org/10.52842/conf.ecaade.2023.2.431
source	Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 2, Graz, 20-22 September 2023, pp. 431–440
summary	This paper discusses a novel deep learning (DL) framework named Building-GNN, which combines the Graph Neural Network (GNN) and the Recurrent neural network (RNN) to address the challenge of generating a controllable 3D voxel building model. The aim is to enable architects and AI to jointly explore the shape and internal spatial planning of 3D building models, forming a co-design paradigm. While the 3D results of previous DL methods, such as 3DGAN, are challenging to control in detail and meet the constraints and preferences of architects' inputs, Building-GNN allows for reasoning about the complex constraint relationships between each voxel. In Building-GNN, the GNN simulates and learns the graph structure relationship between 3D voxels, and the RNN captures the complex interplaying constraint relationships between voxels. The training set consists of 4000 rule-based generated 3D voxel models labeled with different degrees of masking. The quality of the 3D results is evaluated using metrics such as IoU, Fid, and constraint satisfaction. The results demonstrate that adding RNN enhances the accuracy of 3D model shape and voxel relationship prediction. Building-GNN can perform multi-step rational reasoning to complete the 3D model layout planning in different scenarios based on the architect's precise control and incomplete input.
keywords	Deep learning, Graph Neural Networks, 3D Building Layout, Co-design Recurrent Neural Networks, Multi-step Reasoning
series	eCAADe
email
last changed	2023/12/10 10:49

_id	architectural_intelligence2023_7
id	architectural_intelligence2023_7
authors	Immanuel Koh
year	2023
title	Architectural sampling: three possible preconditions for machine learning architectural forms
doi	https://doi.org/https://doi.org/10.1007/s44223-023-00024-1
source	Architectural Intelligence Journal
summary	“We shape our tools, and thereafter our tools shape us” “We need to learn how to see a parallel universe composed of activations, keypoints , eigenfaces, feature transforms, classifiers and training sets”
series	Architectural Intelligence
email
last changed	2025/01/09 15:00

_id	ijac202321208
id	ijac202321208
authors	Ennemoser, Benjamin; Mayrhofer-Hufnagl, Ingrid
year	2023
title	Design across multi-scale datasets by developing a novel approach to 3DGANs
source	International Journal of Architectural Computing 2023, Vol. 21 - no. 2, 358–373
summary	The development of Generative Adversarial Networks (GANs) has accelerated the research of Artificial Intelligence (AI) in architecture as a generative tool. However, since their initial invention, many versions have been developed that only focus on 2D image datasets for training and images as output. The current state of 3DGAN research has yielded promising results. However, these contributions focus primarily on building mass, extrusion of 2D plans, or the overall shape of objects. In comparison, our newly developed 3DGAN approach, using fully spatial building datasets, demonstrates that unprecedented interconnections across different scales are possible resulting in unconventional spatial configurations. Unlike a traditional design process, based on analyzing only a few precedents (typology) according to the task, by collaborating with the machine we can draw on a significantly wider variety of buildings across multiple typologies. In addition, the dataset was extended beyond the scale of complete buildings and involved building components that define space. Thus, our results achieve a high spatial diversity. A detailed analysis of the results also revealed new hybrid architectural elements illustrating that the machine continued the interconnections of scale since elements were not explicitly part of the dataset, becoming a true design collaborator.
keywords	3D Generative adversarial networks, architectural design, Spatial Interpolations
series	journal
last changed	2024/04/17 14:30

_id	caadria2023_26
id	caadria2023_26
authors	Karsan, Zain
year	2023
title	Desk Mate: A Collaborative Drawing Platform
doi	https://doi.org/10.52842/conf.caadria.2023.2.521
source	Immanuel Koh, Dagmar Reinhardt, Mohammed Makki, Mona Khakhar, Nic Bao (eds.), HUMAN-CENTRIC - Proceedings of the 28th CAADRIA Conference, Ahmedabad, 18-24 March 2023, pp. 521–530
summary	Machine Learning (ML) in architecture is an emerging field with myriad potentials to impact the design process. Despite its many possibilities, ML is typically employed when the design problem is sufficiently defined, and further, is only integrated within software environments. Desk Mate is collaborative drawing machine that can be used early in the design process by coupling tangible tools like pens and trace paper with ML driven feedback and generation. Embedding physical tools that are familiar and intuitive with digital intelligence offers designers new ways of engaging with ML algorithms interactively, potentially changing the way the architectural industry approaches design problems. Desk Mate chains together image retrieval methods from machine vision with generative ML models like variational autoencoders (VAE) and generative adversarial networks (GANS) to react to design sketches as they are drawn. This pipeline allows Desk Mate to iterate through designs with the designer. Thus, Desk Mate demonstrates an interactive platform that collocates designer and machine as creative agents, facilitating drawing with ML driven feedback, potentially accelerating design iteration in the early stages of ideation.
keywords	human machine interaction, machine learning and artificial intelligence, interactive machine learning, robotics and autonomous systems
series	CAADRIA
email
last changed	2023/06/15 23:14

For more results click below: