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 592

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

_id	caadria2020_008
id	caadria2020_008
authors	Wang, Likai, Chen, Kian Wee, Janssen, Patrick and Ji, Guohua
year	2020
title	Enabling Optimisation-based Exploration for Building Massing Design - A Coding-free Evolutionary Building Massing Design Toolkit in Rhino-Grasshopper
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 1, Chulalongkorn University, Bangkok, Thailand, 5-6 August 2020, pp. 255-264
doi	https://doi.org/10.52842/conf.caadria.2020.1.255
summary	This paper presents an evolutionary design toolkit for performance-based building massing design optimisation. The toolkit is aimed to assist architects in exploring a wide range of building massing design alternatives guided by various performance objectives, thereby encouraging architects to incorporate evolutionary design optimisation for enriching design ideation at the outset of the design process. The toolkit is implemented in the Rhino-Grasshopper environment and includes components of a diversity-guided evolutionary algorithm and two pre-defined parametric models capable of generating a wide range of massing designs. The evolutionary algorithm can yield diverse design variants from the optimisation process and present more informative results with higher design differentiation. The pre-defined parametric models require minimal customisation from the architects. By using the toolkit, architects can readily explore high-performing building design with performance-based design optimisation with ease, and the coding-free optimisation workflow also streamlines the design process.
keywords	evolutionary design; building massing design; performance-based design; design process; design exploration
series	CAADRIA
email
last changed	2022/06/07 07:58

_id	caadria2020_126
id	caadria2020_126
authors	Hsiao, Chi-Fu, Lee, Ching-Han, Chen, Chun-Yen and Chang, Teng-Wen
year	2020
title	A Co-existing Interactive Approach to Digital Fabrication Workflow
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 1, Chulalongkorn University, Bangkok, Thailand, 5-6 August 2020, pp. 105-114
doi	https://doi.org/10.52842/conf.caadria.2020.1.105
summary	In recent years, digital fabrication projects have explored how to best present complex spatial patterns. These patterns are generated by a series of function clusters and need to be separated into reasonable working sequences for workers. In the stage between design and fabrication, designers and workers typically spend considerable time communicating with each other and prototyping models in order to understand the complex geometry and joint methods of fabrication works. Through the potential of mixed reality technology, this paper proposes a novel form of co-existing interactive workflow that helps designers understand the morphing status of material composition and assists workers in achieving desired results. We establish this co-existing workflow mechanism as an interface between design and reality that includes a HoloLens display, a parametric algorithm, and gesture control identification. This paper challenges the flexibility between the virtual and reality and the interaction between precise parameters and natural gestures within an automation process.
keywords	Co-existing interactive workflow; Digital fabrication; HoloLens; Digital twin; Prototype
series	CAADRIA
email
last changed	2022/06/07 07:51

_id	caadria2020_258
id	caadria2020_258
authors	Beatricia, Beatricia, Indraprastha, Aswin and Koerniawan, M. Donny
year	2020
title	Revisiting Packing Algorithm - A Strategy for Optimum Net-to Gross Office Design
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 1, Chulalongkorn University, Bangkok, Thailand, 5-6 August 2020, pp. 405-414
doi	https://doi.org/10.52842/conf.caadria.2020.1.405
summary	Net-to-gross efficiency is defined as the ratio of net area to a gross area of a building. Net-to-gross efficiency will determine the quantity of leasable building area. On the other side, the effectiveness of the spatial distribution of a floor plan design must follow the value of net-to-gross efficiency. Therefore in the context of office design, there are two challenges need to be improved: 1) to get an optimum value of efficiency, architects need to assign the amount and size of the office units which can be effectively arranged, and 2) to fulfill high net-to-gross efficiency value that usually set out at minimal 85%. This paper aims to apply the packing algorithm as a strategy to achieve optimum net-to-gross efficiency and generating spatial configuration of office units that fit with the result. Our study experimented with series of models and simulations consisting of three stages that start from finding net-to-gross efficiency, defining office unit profiles based on preferable office space units, and applying the packing algorithm to get an optimum office net-to-gross efficiency. Computational processes using physics engine and optimization solvers have been utilized to generate design layouts that have minimal spatial residues, hence increasing the net-to-gross ratio.
keywords	net-to-gross efficiency; packing algorithm; modular office area; area optimization;
series	CAADRIA
email
last changed	2022/06/07 07:54

_id	acadia20_208p
id	acadia20_208p
authors	Bernier-Lavigne, Samuel
year	2020
title	Object-Field
source	ACADIA 2020: Distributed Proximities / Volume II: Projects [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95253-6]. Online and Global. 24-30 October 2020. edited by M. Yablonina, A. Marcus, S. Doyle, M. del Campo, V. Ago, B. Slocum. 208-213
summary	This project aims to continue the correlative study between two fundamental entities of digital architecture: the object and the field. Following periods of experimentations on the ""field"" (materialization of flows of data through animation), the ""field of objects"" (parametricism), the ""object"" (OOO), we investigate the last possible interaction remaining: the ""object-field,"" by merging the formal characteristics of the object with the structural flow of its internal field. This investigation is achieved by exploring the high-resolution features of 3d printing in the design of autonomous architectural objects expressing materiality through topological optimization. The objects are generated by an iterative process of volumetric reduction, resulting in an ensemble of monoliths. Four of them are selected and analyzed through topological optimization in order to extract their internal fields. Next, a series of high-resolution algorithmic systems translate the structural information into 3d printed materiality. Of the four object-fields, one materializes, close to identical, the result of the optimization, giving the keystone to understanding the others. The second one expresses the structural flow through a 1mm voxel system, informed by the optimization, having the effect of stiffening the structure where it is needed and thus generating a new topography on the object. The last two explore the blur that this high-resolution can paradoxically create, with complete integration of the optimal structure in a transparent monolith. This is achieved by a vertex displacement algorithm, and the dissolution of the formal data of the monolith and the structural flows, through the mereological assembly of simple linear elements. For each object-field, a series of drawings was developed using specific algorithmic procedures derived from the peculiarities of their complex geometry. The drawings aim to catalyze coherence throughout the project, where similarities, hitherto kept apart by the multiple materialities, begin to dialogue.
series	ACADIA
type	project
email
last changed	2021/10/26 08:08

_id	sigradi2020_149
id	sigradi2020_149
authors	Canestrino, Giuseppe; Laura, Greco; Spada, Francesco; Lucente, Roberta
year	2020
title	Generating architectural plan with evolutionary multiobjective optimization algorithms: a benchmark case with an existent construction system
source	SIGraDi 2020 [Proceedings of the 24th Conference of the Iberoamerican Society of Digital Graphics - ISSN: 2318-6968] Online Conference 18 - 20 November 2020, pp. 149-156
summary	In architectural design, evolutionary multiobjective optimization algorithms (EMOA) have found use in numerous practical applications in which qualitative and quantitative aspects can be transformed into fitness functions to be optimized. This paper shows that they can be used in an architectural plan design process that starts from a more traditional approach. The benchmark case uses a novel construction system, called Ac.Ca. Building, with a vast architectural and technological database, arleady validated, to generate architectural plan for a residential towerbuilding with a parametric approach and EMOA. The proposed framework differs from past research because uses spatial units with high level of architectural and tecnological definition.
keywords	Architectural design, Parametric architecture, Performance-driven design, architectural layout, evolutionary multiobjective optimization
series	SIGraDi
email
last changed	2021/07/16 11:48

_id	ecaade2020_503
id	ecaade2020_503
authors	Jansen, Igor and Pi¹tek, £ukasz
year	2020
title	The Evolutionary-algorithm-based Automation of the Initial Stage of Apartment Building Design
source	Werner, L and Koering, D (eds.), Anthropologic: Architecture and Fabrication in the cognitive age - Proceedings of the 38th eCAADe Conference - Volume 2, TU Berlin, Berlin, Germany, 16-18 September 2020, pp. 105-114
doi	https://doi.org/10.52842/conf.ecaade.2020.2.105
summary	The development of information technologies has resulted in a strong return of interest in the concept of automating the design process. Most of the attempts such as works of Hersey and Freedman, Duarte or the PRISM application are based on shape grammars. Another approach is evolutionary simulations in concept creation augmentation such as works of Dogan, Saratsis and Reinhart or Nahara and Terzidis.This study examines to what extent evolutionary algorithms can be used to automate early stages of residential multi-family building architectural design. To facilitate informed decision-making, a tool capable of analysing a building plot and proposing the best fitting building shape was designed and tested with Polish legal regulations taken into consideration.A script generating, analysing, and evolutionally optimising a 3D model of the apartment building, was developed. All models met the basic legal conditions and were optimised by four criteria - view obstruction, insolation, maximal allowed floor area built and building compactness. The script was later used on selected building plots producing thousands of solutions. The best performing solutions were selected and presented together with their calculated parameters.
keywords	genetic algorithm; evolutionary simulation; residential building; design automation
series	eCAADe
email
last changed	2022/06/07 07:52

_id	caadria2020_091
id	caadria2020_091
authors	Ren, Yue and Zheng, Hao
year	2020
title	The Spire of AI - Voxel-based 3D Neural Style Transfer
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. 619-628
doi	https://doi.org/10.52842/conf.caadria.2020.2.619
summary	In the architecture field, humans have mastered various skills for creating unique spatial experiences with unknown interplays between known contents and styles. Meanwhile, machine learning, as a popular tool for mapping different input factors and generating unpredictable outputs, links the similarity of the machine intelligence with the typical form-finding process. Style Transfer, therefore, is widely used in 2D visuals for mixing styles while inspiring the architecture field with new form-finding possibilities. Researchers have applied the algorithm in generating 2D renderings of buildings, limiting the results in 2D pixels rather than real full volume forms. Therefore, this paper aims to develop a voxel-based form generation methodology to extend the 3D architectural application of Style Transfer. Briefly, through cutting the original 3D model into multiple plans and apply them to the 2D style image, the stylized 2D results generated by Style Transfer are then abstracted and filtered as groups of pixel points in space. By adjusting the feature parameters with user customization and replacing pixel points with basic voxelization units, designers can easily recreate the original 3D geometries into different design styles, which proposes an intelligent way of finding new and inspiring 3D forms.
keywords	Form Finding; Machine Learning; Artificial Intelligence; Style Transfer
series	CAADRIA
email
last changed	2022/06/07 07:56

_id	ecaade2020_093
id	ecaade2020_093
authors	Veloso, Pedro and Krishnamurti, Ramesh
year	2020
title	An Academy of Spatial Agents - Generating spatial configurations with deep reinforcement learning
source	Werner, L and Koering, D (eds.), Anthropologic: Architecture and Fabrication in the cognitive age - Proceedings of the 38th eCAADe Conference - Volume 2, TU Berlin, Berlin, Germany, 16-18 September 2020, pp. 191-200
doi	https://doi.org/10.52842/conf.ecaade.2020.2.191
summary	Agent-based models rely on decentralized decision making instantiated in the interactions between agents and the environment. In the context of generative design, agent-based models can enable decentralized geometric modelling, provide partial information about the generative process, and enable fine-grained interaction. However, the existing agent-based models originate from non-architectural problems and it is not straight-forward to adapt them for spatial design. To address this, we introduce a method to create custom spatial agents that can satisfy architectural requirements and support fine-grained interaction using multi-agent deep reinforcement learning (MADRL). We focus on a proof of concept where agents control spatial partitions and interact in an environment (represented as a grid) to satisfy custom goals (shape, area, adjacency, etc.). This approach uses double deep Q-network (DDQN) combined with a dynamic convolutional neural-network (DCNN). We report an experiment where trained agents generalize their knowledge to different settings, consistently explore good spatial configurations, and quickly recover from perturbations in the action selection.
keywords	space planning; agent-based model; interactive generative systems; artificial intelligence; multi-agent deep reinforcement learning
series	eCAADe
email
last changed	2022/06/07 07:58

_id	caadria2020_024
id	caadria2020_024
authors	Zheng, Hao and Ren, Yue
year	2020
title	Architectural Layout Design through Simulated Annealing Algorithm
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 1, Chulalongkorn University, Bangkok, Thailand, 5-6 August 2020, pp. 275-284
doi	https://doi.org/10.52842/conf.caadria.2020.1.275
summary	Simulated Annealing is an artificial intelligence algorithm for finding the optimal solution of a proposition in an ample search space, which is based on the similarity between the physical annealing process of solid materials and the combinatorial optimization problem. In architectural layout design, although architects usually rely on their subjective design concepts to arrange buildings in a site, the judging criteria hidden in their design concepts are understandable. They can be summarized and parameterized as a combination of penalty and reward functions. By defining the functions to evaluate a design plan, then using the simulated annealing algorithm to search the optimal solution, the plan can be optimized and generated automatically. Six penalty and reward functions are proposed with different parameter weights in this article, which become a guideline for architectural layout design, especially for residential area planning. Then the results of several tests are shown, in which the parameter weights are adjusted, and the importance of each function is integrated. Lastly, a recommended weight and "temperature" setting are proposed, and a system of generating architectural layout is invented, which releases architects from building arranging work in an early stage.
keywords	Architectural Layout; Simulated Annealing; Artificial Intelligence; Computational Design
series	CAADRIA
email
last changed	2022/06/07 07:57

_id	acadia20_208
id	acadia20_208
authors	Zheng, Hao; Wang, Xinyu; Qi, Zehua; Sun, Shixuan; Akbarzadeh, Masoud
year	2020
title	Generating and Optimizing a Funicular Arch Floor Structure
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 208-217.
doi	https://doi.org/10.52842/conf.acadia.2020.2.208
summary	In this paper, we propose a geometry-based generative design method to generate and optimize a floor structure with funicular building members. This method challenges the antiquated column system, which has been used for more than a century. By inputting the floor plan with the positions of columns, designers can generate a variety of funicular supporting structures, expanding the choice of floor structure designs beyond simply columns and beams and encouraging the creation of architectural spaces with more diverse design elements. We further apply machine learning techniques (artificial neural networks) to evaluate and optimize the structural performance and constructability of the funicular structure, thus finding the optimal solutions within the almost infinite solution space. To achieve this, a machine learning model is trained and used as a fast evaluator to help the evolutionary algorithm find the optimal designs. This interdisciplinary method combines computer science and structural design, providing flexible design choices for generating floor structures.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	ecaade2020_193
id	ecaade2020_193
authors	Alymani, Abdulrahman, Jabi, Wassim and Corcoran, Padraig
year	2020
title	Machine Learning Methods for Clustering Architectural Precedents - Classifying the relationship between building and ground
source	Werner, L and Koering, D (eds.), Anthropologic: Architecture and Fabrication in the cognitive age - Proceedings of the 38th eCAADe Conference - Volume 1, TU Berlin, Berlin, Germany, 16-18 September 2020, pp. 643-652
doi	https://doi.org/10.52842/conf.ecaade.2020.1.643
summary	Every time an object is built, it creates a relationship with the ground. Architects have a full responsibility to design the building by taking the ground into consideration. In the field of architecture, using data mining to identify any unusual patterns or emergent architectural trends is a nascent area that has yet to be fully explored. Clustering techniques are an essential tool in this process for organising large datasets. In this paper, we propose a novel proof-of-concept workflow that enables a machine learning computer system to cluster aspects of an architect's building design style with respect to how the buildings in question relate to the ground. The experimental workflow in this paper consists of two stages. In the first stage, we use a database system to collect, organise and store several significant architectural precedents. The second stage examines the most well-known unsupervised learning algorithm clustering techniques which are: K-Means, K-Modes and Gaussian Mixture Models. Our experiments demonstrated that the K-means clustering algorithm method achieves a level of accuracy that is higher than other clustering methods. This research points to the potential of AI in helping designers identify the typological and topological characteristics of architectural solutions and place them within the most relevant architectural canons
keywords	Machine Learning; Building and Ground Relationship; Clustering Algorithms; K-means cluster Algorithms
series	eCAADe
email
last changed	2022/06/07 07:54

_id	ijac202018202
id	ijac202018202
authors	Pasquero, Claudia and Marco Poletto
year	2020
title	Bio-digital aesthetics as value system of post-Anthropocene architecture
source	International Journal of Architectural Computing vol. 18 - no. 2, 120-140
summary	It is timely within the Anthropocene era, more than ever before, to search for a non-anthropocentric mode of reasoning, and consequently designing. The PhotoSynthetica Consortium, established in 2018 and including London-based ecoLogicStudio, the Urban Morphogenesis Lab (Bartlett School of Architecture, University College London) and the Synthetic Landscape Lab (University of Innsbruck, Austria), has therefore been pursuing architecture as a research-based practice, exploring the interdependence of digital and biological intelligence in design by working directly with non-human living organisms. The research focuses on the diagrammatic capacity of these organisms in the process of growing and becoming part of complex bio-digital architectures. A key remit is training architects’ sensibility at recognising patterns of reasoning across disciplines, materialities and technological regimes, thus expanding the practice’s repertoire of aesthetic qualities. Recent developments in evolutionary psychology demonstrate that the human sense of beauty and pleasure is part of a co-evolutionary system of mind and surrounding environment. In these terms, human senses of beauty and pleasure have evolved as selection mechanisms. Cultivating and enhancing them compensate and integrate the functions of logical thinking to gain a systemic view on the planet Earth and the dramatic changes it is currently undergoing. This article seeks to illustrate, through a series of recent research projects, how a renewed appreciation of beauty in architecture has evolved into an operational tool to design and measure its actual ecological intelligence.
keywords	Bio-digital, bio-computation, bio-city, effectiveness, empathy, impact, sensing
series	journal
email
last changed	2020/11/02 13:34

_id	caadria2020_094
id	caadria2020_094
authors	Yang, Chunxia and Gu, Zhuoxing
year	2020
title	Optimization of Public Space Design Based on Reconstruction of Digital Multi-Agent Behavior - --Taking the public space of the North Bund in Shanghai as an example
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 1, Chulalongkorn University, Bangkok, Thailand, 5-6 August 2020, pp. 335-344
doi	https://doi.org/10.52842/conf.caadria.2020.1.335
summary	This paper uses the digital software platform to build an intelligent multi-agent system. Through the classification of site elements, the Shanghai North Bund waterfront public space elements are classified into different systems such as transportation hub facilities, catering facilities, shopping facilities and leisure venues. The main population activities in this area are classified into different activities such as youth activities, elderly activities, and family activities through user behavior classification. Finally, the intelligent multi-agent particle swarm is built by the dynamic simulation component of grasshopper, and its individual behavior rules and group interaction rules are adjusted to form the crowd moving particle flow. The particle flow interacts with the classified site elements to derive a distribution pattern of population activity in different systems. Particle flow data information and particle distribution patterns after interactive simulation can be the support for urban design evaluation and optimization.
keywords	Self-organizing system; Multi-agent system; Particle property construction; Urban design elements; Waterfront public space
series	CAADRIA
email
last changed	2022/06/07 07:57

_id	acadia20_188p
id	acadia20_188p
authors	Puckett , Nick
year	2020
title	Pulse V2
source	ACADIA 2020: Distributed Proximities / Volume II: Projects [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95253-6]. Online and Global. 24-30 October 2020. edited by M. Yablonina, A. Marcus, S. Doyle, M. del Campo, V. Ago, B. Slocum. 188-191
summary	Pulse v2 is an interactive installation designed to investigate how real-time lidar data can be used to develop new spatial relationships between people and an autonomous digital agent through dynamic visual expressions. The first iteration of this research, Pulse v1, used a single point lidar with a 160o FOV in conjunction with 240 servo-actuated antennas that visualized the position and movement of visitors via their vibrations. This second iteration blends digital and physical materiality to create a synthetic organism that fully integrates sensing, computation, and response into its form. Simultaneously, the raw data feed it “sees” is projected onto the wall in real-time, allowing visitors to experience both the response and the logic. The data feed is supplied by a 360o FOV, 2d lidar scanner. This type of scanner is typically used by small autonomous robots to map and navigate their environments. However, in this installation, the relationship is inverted to allow a stationary agent to respond to a dynamically changing environment. The sensor is mounted under the displays and provides a real-time slice of the space at the height of 20cm. An algorithm filters this data stream into trackable blobs by recognizing people via their ankles. The agent analyzes this stream of data and filters it through a series of micro and macro expressions that play out on the screen in the form of a digital microorganism.
series	ACADIA
type	project
email
last changed	2021/10/26 08:08

_id	ecaade2020_258
id	ecaade2020_258
authors	Zhang, Ran, Waibel, Christoph and Wortmann, Thomas
year	2020
title	Aerodynamic Shape Optimization for High-Rise Conceptual Design - Integrating and validating parametric design, (fast) fluid dynamics, structural analysis and optimization
source	Werner, L and Koering, D (eds.), Anthropologic: Architecture and Fabrication in the cognitive age - Proceedings of the 38th eCAADe Conference - Volume 1, TU Berlin, Berlin, Germany, 16-18 September 2020, pp. 37-45
doi	https://doi.org/10.52842/conf.ecaade.2020.1.037
summary	Using an integrated workflow with parametric design, Computational Fluid Dynamic (CFD) and Fast Fluid Dynamic (FFD) simulations, structural analysis and optimization, this paper evaluates the relative suitability of CFD and FFD simulations for Aerodynamic Shape Optimization (ASO). Specifically, it applies RBFOpt, a model-based optimization algorithm, to the ASO of a supertall high-rise. The paper evaluates the accuracy of the CFD and FDD simulations relative to a slower, more exact CFD simulation, and the performance of the model-based optimization algorithm relative to CMA-ES, an evolutionary algorithm. We conclude that FFD is useful for relative comparisons, such as for optimization, but less accurate than CFD in terms of absolute quantities. Although results tend to be similar, CMA-ES performs less well than RBFOpt for both large and small numbers of simulations, and for both CFD and FFD. RBFOpt with FFD emerges as the most suitable method for conceptual design, as it is much faster and only slightly less effective than RBFOpt with CFD.
keywords	Aerodynamic Shape Optimization; Computational Fluid Dynamics (CFD); Fast Fluid Dynamics (FFD); Model-based Optimization; High-rise Conceptual Design
series	eCAADe
email
last changed	2022/06/07 07:57

_id	acadia20_228
id	acadia20_228
authors	Alawadhi, Mohammad; Yan, Wei
year	2020
title	BIM Hyperreality
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 228-236.
doi	https://doi.org/10.52842/conf.acadia.2020.1.228
summary	Deep learning is expected to offer new opportunities and a new paradigm for the field of architecture. One such opportunity is teaching neural networks to visually understand architectural elements from the built environment. However, the availability of large training datasets is one of the biggest limitations of neural networks. Also, the vast majority of training data for visual recognition tasks is annotated by humans. In order to resolve this bottleneck, we present a concept of a hybrid system—using both building information modeling (BIM) and hyperrealistic (photorealistic) rendering—to synthesize datasets for training a neural network for building object recognition in photos. For generating our training dataset, BIMrAI, we used an existing BIM model and a corresponding photorealistically rendered model of the same building. We created methods for using renderings to train a deep learning model, trained a generative adversarial network (GAN) model using these methods, and tested the output model on real-world photos. For the specific case study presented in this paper, our results show that a neural network trained with synthetic data (i.e., photorealistic renderings and BIM-based semantic labels) can be used to identify building objects from photos without using photos in the training data. Future work can enhance the presented methods using available BIM models and renderings for more generalized mapping and description of photographed built environments.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	acadia20_120
id	acadia20_120
authors	Barsan-Pipu, Claudiu; Sleiman, Nathalie; Moldovan, Theodor
year	2020
title	Affective Computing for Generating Virtual Procedural Environments Using Game Technologies
source	ACADIA 2020: Distributed Proximities / Volume I: Technical Papers [Proceedings of the 40th Annual Conference of the Association of Computer Aided Design in Architecture (ACADIA) ISBN 978-0-578-95213-0]. Online and Global. 24-30 October 2020. edited by B. Slocum, V. Ago, S. Doyle, A. Marcus, M. Yablonina, and M. del Campo. 120-129.
doi	https://doi.org/10.52842/conf.acadia.2020.2.120
summary	Architects have long sought to create spaces that can relate to or even induce specific emotional conditions in their users, such as states of relaxation or engagement. Dynamic or calming qualities were given to these spaces by controlling form, perspective, lighting, color, and materiality. The actual impact of these complex design decisions has been challenging to assess, from both quantitative and qualitative standpoints, because neural empathic responses, defined in this paper by feature indexes (FIs) and mind indexes (MIs), are highly subjective experiences. Recent advances in the fields of virtual procedural environments (VPEs) and virtual reality (VR), supported by powerful game engine (GE) technologies, provide computational designers with a new set of design instruments that, when combined with brain-computing interfacing (BCI) and eye-tracking (E-T) hardware, can be used to assess complex empathic reactions. As the COVID-19 health crisis showed, virtual social interaction becomes increasingly relevant, and the social catalytic potential of VPEs can open new design possibilities. The research presented in this paper introduces the cyber-physical design of such an affective computing system. It focuses on how relevant empathic data can be acquired in real time by exposing subjects within a dynamic VR-based VPE and assessing their emotional responses while controlling the actual generative parameters via a live feedback loop. A combination of VR, BCI, and E-T solutions integrated within a GE is proposed and discussed. By using a VPE inside a BCI system that can be accurately correlated with E-T, this paper proposes to identify potential morphological and lighting factors that either alone or combined can have an empathic effect expressed by the relevant responses of the MIs.
series	ACADIA
type	paper
email
last changed	2023/10/22 12:06

_id	cdrf2019_17
id	cdrf2019_17
authors	Chuan Liu, Jiaqi Shen, Yue Ren, and Hao Zheng
year	2020
title	Pipes of AI – Machine Learning Assisted 3D Modeling Design
source	Proceedings of the 2020 DigitalFUTURES The 2nd International Conference on Computational Design and Robotic Fabrication (CDRF 2020)
doi	https://doi.org/https://doi.org/10.1007/978-981-33-4400-6_2
summary	Style transfer is a design technique that is based on Artificial Intelligence and Machine Learning, which is an innovative way to generate new images with the intervention of style images. The output image will carry the characteristic of style image and maintain the content of the input image. However, the design technique is employed in generating 2D images, which has a limited range in practical use. Thus, the goal of the project is to utilize style transfer as a toolset for architectural design and find out the possibility for a 3D modeling design. To implement style transfer into the research, floor plans of different heights are selected from a given design boundary and set as the content images, while a framework of a truss structure is set as the style image. Transferred images are obtained after processing the style transfer neural network, then the geometric images are translated into floor plans for new structure design. After the selection of the tilt angle and the degree of density, vertical components that connecting two adjacent layers are generated to be the pillars of the structure. At this stage, 2D style transferred images are successfully transformed into 3D geometries, which can be applied to the architectural design processes. Generally speaking, style transfer is an intelligent design tool that provides architects with a variety of choices of idea-generating. It has the potential to inspire architects at an early stage of design with not only 2D but also 3D format.
series	cdrf
email
last changed	2022/09/29 07:51

_id	ecaade2021_257
id	ecaade2021_257
authors	Cichocka, Judyta Maria, Loj, Szymon and Wloczyk, Marta Magdalena
year	2021
title	A Method for Generating Regular Grid Configurations on Free-From Surfaces for Structurally Sound Geodesic Gridshells
source	Stojakovic, V and Tepavcevic, B (eds.), Towards a new, configurable architecture - Proceedings of the 39th eCAADe Conference - Volume 2, University of Novi Sad, Novi Sad, Serbia, 8-10 September 2021, pp. 493-502
doi	https://doi.org/10.52842/conf.ecaade.2021.2.493
summary	Gridshells are highly efficient, lightweight structures which can span long distances with minimal use of material (Vassallo & Malek 2017). One of the most promising and novel categories of gridshells are bending-active (elastic) systems (Lienhard & Gengnagel 2018), which are composed of flexible members (Kuijenhoven & Hoogenboom 2012). Timber elastic gridshells can be site-sprung or sequentially erected (geodesic). While a lot of research focus is on the site-sprung ones, the methods for design of sequentially-erected geodesic gridshells remained underdeveloped (Cichocka 2020). The main objective of the paper is to introduce a method of generating regular geodesic grid patterns on free-form surfaces and to examine its applicability to design structurally feasible geodesic gridshells. We adopted differential geometry methods of generating regular bidirectional geodesic grids on free-form surfaces. Then, we compared the structural performance of the regular and the irregular grids of the same density on three free-form surfaces. The proposed method successfully produces the regular geodesic grid patterns on the free-form surfaces with varying curvature-richness. Our analysis shows that gridshells with regular grid configurations perform structurally better than those with irregular patterns. We conclude that the presented method can be readily used and can expand possibilities of application of geodesic gridshells.
keywords	elastic timber gridshell; bending-active structure; grid configuration optimization; computational differential geometry; material-based design methodology; free-form surface; pattern; geodesic
series	eCAADe
email
last changed	2022/06/07 07:56

_id	ecaade2020_222
id	ecaade2020_222
authors	Ikeno, Kazunosuke, Fukuda, Tomohiro and Yabuki, Nobuyoshi
year	2020
title	Automatic Generation of Horizontal Building Mask Images by Using a 3D Model with Aerial Photographs for Deep Learning
source	Werner, L and Koering, D (eds.), Anthropologic: Architecture and Fabrication in the cognitive age - Proceedings of the 38th eCAADe Conference - Volume 2, TU Berlin, Berlin, Germany, 16-18 September 2020, pp. 271-278
doi	https://doi.org/10.52842/conf.ecaade.2020.2.271
summary	Information extracted from aerial photographs is widely used in urban planning and design. An effective method for detecting buildings in aerial photographs is to use deep learning for understanding the current state of a target region. However, the building mask images used to train the deep learning model are manually generated in many cases. To solve this challenge, a method has been proposed for automatically generating mask images by using virtual reality 3D models for deep learning. Because normal virtual models do not have the realism of a photograph, it is difficult to obtain highly accurate detection results in the real world even if the images are used for deep learning training. Therefore, the objective of this research is to propose a method for automatically generating building mask images by using 3D models with textured aerial photographs for deep learning. The model trained on datasets generated by the proposed method could detect buildings in aerial photographs with an accuracy of IoU = 0.622. Work left for the future includes changing the size and type of mask images, training the model, and evaluating the accuracy of the trained model.
keywords	Urban planning and design; Deep learning; Semantic segmentation; Mask image; Training data; Automatic design
series	eCAADe
email
last changed	2022/06/07 07:50

For more results click below: