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 613

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

_id	caadria2018_044
id	caadria2018_044
authors	Inoue, Kazuya, Fukuda, Tomohiro, Cao, Rui and Yabuki, Nobuyoshi
year	2018
title	Tracking Robustness and Green View Index Estimation of Augmented and Diminished Reality for Environmental Design - PhotoAR+DR2017 project
doi	https://doi.org/10.52842/conf.caadria.2018.1.339
source	T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 1, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 339-348
summary	To assess an environmental design, augmented and diminished reality (AR/DR) have a potential to build a consensus more smoothly through the landscape simulation of new design visualization of the items to be assessed, such as the green view index. However, the current system is still considered to be impractical because it does not provide complete user experience. Thus, we aim to improve the robustness of the AR/DR system and to integrate the estimation of the green view index into the AR/DR system on a game engine. Further, we achieve an improved stable tracking by eliminating the outliers of the tracking reference points using the random sample consensus (RANSAC) method and by defining the tracking reference points over an extensive area of the AR/DR display. Additionally, two modules were implemented, among which one module is used to solve the occlusion problem while the other is used to estimate the green view index. The novel integrated AR/DR system with all modules was developed on the game engine. A mock design project was developed in an outdoor environment for simulation purposes, thereby verifying the applicability of the developed system.
keywords	Environmental Design; Augmented Reality (AR); Diminished Reality (DR); Green View Index; Segmentation
series	CAADRIA
email
last changed	2022/06/07 07:50

_id	caadria2018_046
id	caadria2018_046
authors	Lu, Siliang and Cochran Hameen, Erica
year	2018
title	Integrated IR Vision Sensor for Online Clothing Insulation Measurement
doi	https://doi.org/10.52842/conf.caadria.2018.1.565
source	T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 1, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 565-573
summary	As one of the most important building systems, HVAC plays a key role in creating a comfortable thermal environment. Predicted Mean Vote (PMV), an index that predicts the mean value of the votes of a large group of persons on the thermal sensation scale, has been adopted to evaluate the built environment. Compared to environmental factors, clothing insulation can be much harder to measure in the field. The existing research on real-time clothing insulation measurement mainly focuses on expensive infrared thermography (IR) cameras. Therefore, to ensure cost-effectiveness, the paper has proposed a solution consisting of a normal camera, IR and air temperature sensors and Arduino Nanos to measure clothing insulation in real-time. Moreover, the algorithm includes the initialization from clothing classification with pre-trained neural network and optimization of the clothing insulation calculation. A total of 8 tests have been conducted with garments for spring/fall, summer and winter. The current results have shown the accuracy of T-shirt classification can reach over 90%. Moreover, compared with the results with IR cameras and reference values, the accuracies of the proposed sensing system vary with different clothing types. Research shall be further conducted and be applied into the dynamic PMV-based HVAC control system.
keywords	clothing insulation; skin temperature; clothing classification; IR temperature sensor; Optimization
series	CAADRIA
email
last changed	2022/06/07 07:59

_id	caadria2018_314
id	caadria2018_314
authors	Kim, Jin Sung, Song, Jae Yeol and Lee, Jin Kook
year	2018
title	Approach to the Extraction of Design Features of Interior Design Elements Using Image Recognition Technique
doi	https://doi.org/10.52842/conf.caadria.2018.2.287
source	T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 2, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 287-296
summary	This paper aims to propose deep learning-based approach to the auto-recognition of their design features of interior design elements using given digital images. The recently image recognition technique using convolutional neural networks has shown great success in the various field of research and industry. The open-source frameworks and pre-trained image recognition models supporting image recognition task enable us to easily retrain the models to apply them on any domain. This paper describes how to apply such techniques on interior design process and depicts some demonstration results in that approaches. Furniture that is one of the most common interior design elements has sub-feature including implicit design features, such as style, shape, function as well as explicit properties, such as component, materials, and size. This paper shows to retrain the model to extract some of the features for efficiently managing and utilizing such design information. The target element is chair and the target design features are limited to functional features, materials, seating capacity and design style. Total 3933 chair images dataset and 6 retrained image recognition models were utilized for retraining. Through the combination of those multiple models, inference demonstration also has been described.
keywords	Deep learning; Image recognition; Interior design elements; Design feature; Chair
series	CAADRIA
email
last changed	2022/06/07 07:52

_id	acadia18_404
id	acadia18_404
authors	Clifford, Brandon; McGee, Wes
year	2018
title	Cyclopean Cannibalism. A method for recycling rubble
doi	https://doi.org/10.52842/conf.acadia.2018.404
source	ACADIA // 2018: Recalibration. On imprecisionand infidelity. [Proceedings of the 38th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-692-17729-7] Mexico City, Mexico 18-20 October, 2018, pp. 404-413
summary	Each year, the United States discards 375 million tons of concrete construction debris to landfills (U.S. EPA 2016), but this is a new paradigm. Past civilizations cannibalized their constructions to produce new architectures (Hopkins 2005). This paper interrogates one cannibalistic methodology from the past known as cyclopean masonry in order to translate this valuable method into a contemporary digital procedure. The work contextualizes the techniques of this method and situates them into procedural recipes which can be applied in contemporary construction. A full-scale prototype is produced utilizing the described method; demolition debris is gathered, scanned, and processed through an algorithmic workflow. Each rubble unit is then minimally carved by a robotic arm and set to compose a new architecture from discarded rubble debris. The prototype merges ancient construction thinking with digital design and fabrication methodologies. It poses material cannibalism as a means of combating excessive construction waste generation.
keywords	full paper, cyclopean, algorithmic, robotic fabrication, stone, shape grammars, computation
series	ACADIA
type	paper
email
last changed	2022/06/07 07:56

_id	ecaade2018_145
id	ecaade2018_145
authors	Fukuda, Tomohiro, Zhu, Yuehan and Yabuki, Nobuyoshi
year	2018
title	Point Cloud Stream on Spatial Mixed Reality - Toward Telepresence in Architectural Field
doi	https://doi.org/10.52842/conf.ecaade.2018.2.727
source	Kepczynska-Walczak, A, Bialkowski, S (eds.), Computing for a better tomorrow - Proceedings of the 36th eCAADe Conference - Volume 2, Lodz University of Technology, Lodz, Poland, 19-21 September 2018, pp. 727-734
summary	In remote meetings that involve the study of buildings and cities, sharing three-dimensional (3D) virtual spatial of buildings and cities is just as necessary as sharing the appearances and voices of meeting participants. Because of this, system development and pilot projects have attempted to share 3D virtual models via the internet in real-time but is still insufficient compared with face-to-face meeting. Therefore, this research explores the applicability of a spatial mixed reality (MR) system that displays point cloud streams to realize 3D remote meeting in architecture and urban fields. MR is a new technology that enables 3D presentations of various information, combining the physical and virtual worlds. One MR method is telepresence, which is expected to give people a way to communicate remotely as if face to face in a realistic way. We first developed a MR system named PcsMR (Point cloud stream on mixed reality) to display point cloud streams. The PcsMR system's operation consists of generating and transferring a point cloud stream and then rendering a point cloud stream using MR. The PcsMR acquired the point cloud stream in real-time using Kinect for Windows v2 and transferred it to Microsoft HoloLens, which uses optical see-through MR. Then we constructed two prototypes based on PcsMR and carried out pilot projects. Through observing the experiments, application possibilities for architecture and urban fields are found in meetings and communications that share real-time 3D objects and include the movement of remote participants and objects. The proposed method was evaluated feasible and effective.
keywords	Telepresence; Mixed reality; Point cloud stream; Remote meeting; Real time
series	eCAADe
email
last changed	2022/06/07 07:50

_id	caadria2018_268
id	caadria2018_268
authors	Lim, Joie, Janssen, Patrick and Stouffs, Rudi
year	2018
title	Automated Generation of BIM Models from 2D CAD Drawings
doi	https://doi.org/10.52842/conf.caadria.2018.2.061
source	T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 2, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 61-70
summary	Existing buildings are often lacking BIM models. This paper proposes a method to semi-automate the generation of BIM models from 2D CAD drawings. The method has two parts: the first part, 2D CAD drawing preparation, involves cleaning the drawings to obtain simplified 2D input geometry and the second, 3D BIM model generation, involves generating and extracting parameters to generate 3D BIM components. This research focuses on the semi-automation of the second part. The the model is generated storey by storey, with each building element type being processed. A demonstration was carried out for a case-study building. The main modelling strategies used by the method are described and key challenges are discussed.
keywords	BIM; CAD drawings; conversion; generation; Grasshopper
series	CAADRIA
email
last changed	2022/06/07 07:59

_id	ecaade2023_10
id	ecaade2023_10
authors	Sepúlveda, Abel, Eslamirad, Nasim and De Luca, Francesco
year	2023
title	Machine Learning Approach versus Prediction Formulas to Design Healthy Dwellings in a Cold Climate
doi	https://doi.org/10.52842/conf.ecaade.2023.2.359
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. 359–368
summary	This paper presents a study about the prediction accuracy of daylight provision and overheating levels in dwellings when considering different methods (machine learning vs prediction formulas), training, and validation data sets. An existing high-rise building located in Tallinn, Estonia was considered to compare the best ML predictive method with novel prediction formulas. The quantification of daylight provision was conducted according to the European daylight standard EN 17037:2018 (based on minimum Daylight Factor (minDF)) and overheating level in terms of the degree-hour (DH) metric included in local regulations. The features included in the dataset are the minDF and DH values related to different combinations of design parameters: window-to-floor ratio, level of obstruction, g-value, and visible transmittance of the glazing system. Different training and validation data sets were obtained from a main data set of 5120 minDF values and 40960 DH values obtained through simulation with Radiance and EnergyPlus, respectively. For each combination of training and validation dataset, the accuracy of the ML model was quantified and compared with the accuracy of the prediction formulas. According to our results, the ML model could provide more accurate minDF/DH predictions than by using the prediction formulas for the same design parameters. However, the amount of room combinations needed to train the machine-learning model is larger than for the calibration of the prediction formulas. The paper discuss in detail the method to use in practice, depending on time and accuracy concerns.
keywords	Optimization, Daylight, Thermal Comfort, Overheating, Machine Learning, Predictive Model, Dwellings, Cold Climates
series	eCAADe
email
last changed	2023/12/10 10:49

_id	ecaade2020_445
id	ecaade2020_445
authors	Spiegelhalter, Thomas, Andia, Alfredo, Levente, Juhasz and Namuduri, Srikanth
year	2020
title	Part 1: The Integrated Decision Support System - Generative and synthetic biological design imaginations for the Miami bay area
doi	https://doi.org/10.52842/conf.ecaade.2020.2.011
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. 11-20
summary	In less than 150 years our carbon society transformed the planet. Today more than 50% of ecologies in the world are determined by unsustainable industrialization processes. The latest IPCC reports show that we are quickly arriving at points of no return in the warming of our planet. We cannot afford to continue in the same direction, we need a new imagination. As part of an E.U.-US funded $1.9 million research project we have been working on multiple projects for the future of the Miami islands since 2018:1. We developed a generative GIS-BIM based Python API for mapping and optimization of carbon-neutral design workflows. It includes genetic design combinatorics with intuitive graphical Dynamo-Python-Grasshopper programming with experimental design results. 2. We worked on a series of design research for the Miami Bay that envisions islands, living shorelines, programmable soils, and infrastructures that grow by themselves using synthetic biology.
keywords	Automated Workflows, Synthetic Biology, Artificial Intelligence, Architecture, Sea-level Rise
series	eCAADe
email
last changed	2022/06/07 07:56

_id	ecaade2018_243
id	ecaade2018_243
authors	Gardner, Nicole
year	2018
title	Architecture-Human-Machine (re)configurations - Examining computational design in practice
doi	https://doi.org/10.52842/conf.ecaade.2018.2.139
source	Kepczynska-Walczak, A, Bialkowski, S (eds.), Computing for a better tomorrow - Proceedings of the 36th eCAADe Conference - Volume 2, Lodz University of Technology, Lodz, Poland, 19-21 September 2018, pp. 139-148
summary	This paper outlines a research project that explores the participation in, and perception of, advanced technologies in architectural professional practice through a sociotechnical lens and presents empirical research findings from an online survey distributed to employees in five large-scale architectural practices in Sydney, Australia. This argues that while the computational design paradigm might be well accepted, understood, and documented in academic research contexts, the extent and ways that computational design thinking and methods are put-into-practice has to date been less explored. In engineering and construction, technology adoption studies since the mid 1990s have measured information technology (IT) use (Howard et al. 1998; Samuelson and Björk 2013). In architecture, research has also focused on quantifying IT use (Cichocka 2017), as well as the examination of specific practices such as building information modelling (BIM) (Cardoso Llach 2017; Herr and Fischer 2017; Son et al. 2015). With the notable exceptions of Daniel Cardoso Llach (2015; 2017) and Yanni Loukissas (2012), few scholars have explored advanced technologies in architectural practice from a sociotechnical perspective. This paper argues that a sociotechnical lens can net valuable insights into advanced technology engagement to inform pedagogical approaches in architectural education as well as strategies for continuing professional development.
keywords	Computational design; Sociotechnical system; Technology adoption
series	eCAADe
email
last changed	2022/06/07 07:51

_id	cf2019_003
id	cf2019_003
authors	Steinfeld, Kyle; Katherine Park, Adam Menges and Samantha Walker
year	2019
title	Fresh Eyes A framework for the application of machine learning to generative architectural design, and a report of activities at Smartgeometry 2018
source	Ji-Hyun Lee (Eds.) "Hello, Culture!" [18th International Conference, CAAD Futures 2019, Proceedings / ISBN 978-89-89453-05-5] Daejeon, Korea, p. 22
summary	This paper presents a framework for the application of Machine Learning (ML) to Generative Architectural Design (GAD), and illustrates this framework through a description of a series of projects completed at the Smart Geometry conference in May of 2018 (SG 2018) in Toronto. Proposed here is a modest modification of a 3-step process that is well-known in generative architectural design, and that proceeds as: generate, evaluate, iterate. In place of the typical approaches to the evaluation step, we propose to employ a machine learning process: a neural net trained to perform image classification. This modified process is different enough from traditional methods as to warrant an adjustment of the terms of GAD. Through the development of this framework, we seek to demonstrate that generative evaluation may be seen as a new locus of subjectivity in design.
keywords	Machine Learning, Generative Design, Design Methods
series	CAAD Futures
email
last changed	2019/07/29 14:08

_id	cdrf2021_286
id	cdrf2021_286
authors	Yimeng Wei, Areti Markopoulou, Yuanshuang Zhu,Eduardo Chamorro Martin, and Nikol Kirova
year	2021
title	Additive Manufacture of Cellulose Based Bio-Material on Architectural Scale
doi	https://doi.org/https://doi.org/10.1007/978-981-16-5983-6_27
source	Proceedings of the 2021 DigitalFUTURES The 3rd International Conference on Computational Design and Robotic Fabrication (CDRF 2021)
summary	There are severe environmental and ecological issues once we evaluate the architecture industry with LCA (Life Cycle Assessment), such as emission of CO2 caused by necessary high temperature for producing cement and significant amounts of Construction Demolition Waste (CDW) in deteriorated and obsolete buildings. One of the ways to solve these problems is Bio-Material. CELLULOSE and CHITON is the 1st and 2nd abundant substance in nature (Duro-Royo, J.: Aguahoja_ProgrammableWater-based Biocomposites for Digital Design and Fabrication across Scales. MIT, pp. 1–3 (2019)), which means significantly potential for architectural dimension production. Meanwhile, renewability and biodegradability make it more conducive to the current problem of construction pollution. The purpose of this study is to explore Cellulose Based Biomaterial and bring it into architectural scale additive manufacture that engages with performance in the material development, with respect to time of solidification and control of shrinkage, as well as offering mechanical strength. At present, the experiments have proved the possibility of developing a cellulose-chitosan- based composite into 3D-Printing Construction Material (Sanandiya, N.D., Vijay, Y., Dimopoulou, M., Dritsas, S., Fernandez, J.G.: Large-scale additive manufacturing with bioinspired cellulosic materials. Sci. Rep. 8(1), 1–5 (2018)). Moreover, The research shows that the characteristics (Such as waterproof, bending, compression, tensile, transparency) of the composite can be enhanced by different additives (such as xanthan gum, paper fiber, flour), which means it can be customized into various architectural components based on Performance Directional Optimization. This solution has a positive effect on environmental impact reduction and is of great significance in putting the architectural construction industry into a more environment-friendly and smart state.
series	cdrf
email
last changed	2022/09/29 07:53

_id	ijac201816406
id	ijac201816406
authors	As, Imdat; Siddharth Pal and Prithwish Basu
year	2018
title	Artificial intelligence in architecture: Generating conceptual design via deep learning
source	International Journal of Architectural Computing vol. 16 - no. 4, 306-327
summary	Artificial intelligence, and in particular machine learning, is a fast-emerging field. Research on artificial intelligence focuses mainly on image-, text- and voice-based applications, leading to breakthrough developments in self-driving cars, voice recognition algorithms and recommendation systems. In this article, we present the research of an alternative graph- based machine learning system that deals with three-dimensional space, which is more structured and combinatorial than images, text or voice. Specifically, we present a function-driven deep learning approach to generate conceptual design. We trained and used deep neural networks to evaluate existing designs encoded as graphs, extract significant building blocks as subgraphs and merge them into new compositions. Finally, we explored the application of generative adversarial networks to generate entirely new and unique designs.
keywords	Architectural design, conceptual design, deep learning, artificial intelligence, generative design
series	journal
email
last changed	2019/08/07 14:04

_id	ecaade2018_377
id	ecaade2018_377
authors	Beaudry Marchand, Emmanuel, Dorta, Tomás and Pierini, Davide
year	2018
title	Influence of Immersive Contextual Environments on Collaborative Ideation Cognition - Through design conversations, gestures and sketches
doi	https://doi.org/10.52842/conf.ecaade.2018.2.795
source	Kepczynska-Walczak, A, Bialkowski, S (eds.), Computing for a better tomorrow - Proceedings of the 36th eCAADe Conference - Volume 2, Lodz University of Technology, Lodz, Poland, 19-21 September 2018, pp. 795-804
summary	In the design studio, Virtual Reality (VR) has mainly been included as a visualization tool to explore pre-designed ideas developed in traditional 3D software or using pen on paper. Meanwhile, a reshaping of the design process has been taking place, bringing forward interaction/experiential concerns and co-design approaches throughout disciplines in a push for a more thorough consideration of projects' contexts. This paper reports an exploratory study of how immersive contextual representations influence the co-ideation process. Audio-video recordings of co-ideation sessions (9) from a pedagogical studio were analyzed through verbal and representational (sketches and design gestures) exchanges as occurring in three different conditions: (a) pen on paper, immersive headset-free VR (b) without, and (c) with the use of contextual immersive environment (photogrammetric scans and 3D models). Results show that, although design conversations were similar across all conditions, design gestures were more often directly related to- than independent from the graphical representation only when using an immersive contextual environment. Furthermore, the rate of sketching episodes in general and sketching explanations were considerably lower in this condition. This could imply that use of pre-made context greatly reduces the need of sketching elements to support a clearer co-ideation.
keywords	Immersive context; Design gestures; Design conversations; Sketches; Co-design studio; Design cognition
series	eCAADe
email
last changed	2022/06/07 07:54

_id	ecaade2018_329
id	ecaade2018_329
authors	De Luca, Francesco, Nejur, Andrei and Dogan, Timur
year	2018
title	Facade-Floor-Cluster - Methodology for Determining Optimal Building Clusters for Solar Access and Floor Plan Layout in Urban Environments
doi	https://doi.org/10.52842/conf.ecaade.2018.2.585
source	Kepczynska-Walczak, A, Bialkowski, S (eds.), Computing for a better tomorrow - Proceedings of the 36th eCAADe Conference - Volume 2, Lodz University of Technology, Lodz, Poland, 19-21 September 2018, pp. 585-594
summary	Daylight standards are one of the main factors for the shape and image of cities. With urbanization and ongoing densification of cities, new planning regulations are emerging in order to manage access to sun light. In Estonia a daylight standard defines the rights of light for existing buildings and the direct solar access requirement for new premises. The solar envelope method and environmental simulations to compute direct sun light hours on building façades can be used to design buildings that respect both daylight requirements. However, no existing tool integrates both methods in an easy to use manner. Further, the assessment of façade performance needs to be related to the design of interior layouts and of building clusters to be meaningful to architects. Hence, the present work presents a computational design workflow for the evaluation and optimisation of high density building clusters in urban environments in relation to direct solar access requirements and selected types of floor plans.
keywords	Performance-driven Design; Urban Design; Direct Solar Access; Environmental Simulations and Evaluations; Parametric Modelling
series	eCAADe
email
last changed	2022/06/07 07:55

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

_id	sigradi2018_1875
id	sigradi2018_1875
authors	Kalantari, Cruze-Garza; Banner, Pamela; Contreras-Vidal, Jose Luis
year	2018
title	Computationally Analyzing Biometric Data and Virtual Response Testing in Evaluating Learning Performance of Educational Setting Through
source	SIGraDi 2018 [Proceedings of the 22nd Conference of the Iberoamerican Society of Digital Graphics - ISSN: 2318-6968] Brazil, São Carlos 7 - 9 November 2018, pp. 390-396
summary	Due to construction costs, the human effects of innovations in architectural design can be expensive to test. Post-occupancy studies provide valuable data about what did and did not work in the past, but they cannot provide direct feedback for new ideas that have not yet been attempted. This presents designers with something of a dilemma. How can we harness the best potential of new technology and design innovation, while avoiding costly and potentially harmful mistakes? The current research use virtual immersion and biometric data to provide a new form of extremely rigorous human-response testing prior to construction. The researchers’ hypothesis was that virtual test runs can help designers to identify potential problems and successes in their work prior to its being physically constructed. The pilot study aims to develop a digital pre-occupancy toolset to understand the impact of different interior design variables of learning environment (independent variables) on learning performance (dependent variable). This project provides a practical toolset to test the potential human impacts of architectural design innovations. The research responds to a growing call in the field for evidence-based design and for an inexpensive means of evaluating the potential human effects of new designs. Our research will address this challenge by developing a prototype mobile brain-body imaging interface that can be used in conjunction with virtual immersion.
keywords	Signal Processing; Brain; EEG; Virtual Reality; Big Data; Learning Performance
series	SIGRADI
email
last changed	2021/03/28 19:58

_id	acadia18_166
id	acadia18_166
authors	Kvochick, Tyler
year	2018
title	Sneaky Spatial Segmentation. Reading Architectural Drawings with Deep Neural Networks and Without Labeling Data
doi	https://doi.org/10.52842/conf.acadia.2018.166
source	ACADIA // 2018: Recalibration. On imprecisionand infidelity. [Proceedings of the 38th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-692-17729-7] Mexico City, Mexico 18-20 October, 2018, pp. 166-175
summary	Currently, it is nearly impossible for an artificial neural network to generalize a task from very few examples. Humans, however, excel at this. For instance, it is not necessary for a designer to see thousands or millions of unique examples of how to place a given drawing symbol in a way that meets the economic, aesthetic, and performative goals of the project. In fact, the goals can be (and usually are) communicated abstractly in natural language. Machine learning (ML) models, however, do need numerous examples. The methods that we explore here are an attempt to circumvent this in order to make ML models more immediately useful. In this work, we present progress on the application of contemporary ML techniques to the design process in the architecture, engineering, and construction (AEC) industry. We introduce a technique to partially circumvent the data hungriness of neural networks, which is a significant impediment to their application outside of the ML research community. We also show results on the applicability of this technique to real-world drawings and present research that addresses how some fundamental attributes of drawings as images affect the way they are interpreted in deep neural networks. Our primary contribution is a technique to train a neural network to segment real-world architectural drawings after using only generated pseudodrawings.
keywords	full paper, representation + perception, computation, ai & machine learning
series	ACADIA
type	paper
email
last changed	2022/06/07 07:51

_id	ecaade2018_108
id	ecaade2018_108
authors	Luo, Dan, Wang, Jingsong and Xu, Weiguo
year	2018
title	Applied Automatic Machine Learning Process for Material Computation
doi	https://doi.org/10.52842/conf.ecaade.2018.1.109
source	Kepczynska-Walczak, A, Bialkowski, S (eds.), Computing for a better tomorrow - Proceedings of the 36th eCAADe Conference - Volume 1, Lodz University of Technology, Lodz, Poland, 19-21 September 2018, pp. 109-118
summary	Machine learning enables computers to learn without being explicitly programmed. This paper outlines state-of-the-art implementations of machine learning approaches to the study of physical material properties based on Elastomer we developed, which combines with robotic automation and image recognition to generate a computable material model for non-uniform linear Elastomer material. The development of the neural network includes a few preliminary experiments to confirm the feasibility and the influential parameters used to define the final RNN neural network, the study of the inputs and the quality of the testing samples influencing the accuracy of the output model, and the evaluation of the generated material model as well as the method itself. To conclude, this paper expands such methods to the possible architectural implications on other non-uniform materials, such as the performance of wood sheets with different grains and tensile material made from composite materials.
keywords	neural network; robotic; material computation; automation
series	eCAADe
email
last changed	2022/06/07 07:59

_id	caadria2018_083
id	caadria2018_083
authors	Luo, Dan, Wang, Jinsong and Xu, Weiguo
year	2018
title	Robotic Automatic Generation of Performance Model for Non-Uniform Linear Material via Deep Learning
doi	https://doi.org/10.52842/conf.caadria.2018.1.039
source	T. Fukuda, W. Huang, P. Janssen, K. Crolla, S. Alhadidi (eds.), Learning, Adapting and Prototyping - Proceedings of the 23rd CAADRIA Conference - Volume 1, Tsinghua University, Beijing, China, 17-19 May 2018, pp. 39-48
summary	In the following research, a systematic approach is developed to generate an experiment-based performance model that computes and customizes properties of non-uniform linear materials to accommodate the form of designated curve under bending and natural force. In this case, the test subject is an elastomer strip of non-uniform sections. A novel solution is provided to obtain sufficient training data required for deep learning with an automatic material testing mechanism combining robotic arm automation and image recognition. The collected training data are fed into a deep combination of neural networks to generate a material performance model. Unlike most traditional performance models that are only able to simulate the final form from the properties and initial conditions of the given materials, the trained neural network offers a two-way performance model that is also able to compute appropriate material properties of non-uniform materials from target curves. This network achieves complex forms with minimal and effective programmed materials with complicated nonlinear properties and behaving under natural forces.
keywords	Material performance model; Deep Learning; Robotic automation; Material computation; Neural network
series	CAADRIA
email
last changed	2022/06/07 07:59

_id	acadia18_250
id	acadia18_250
authors	Seibold, Zach; Grinham, Jonathan; Geletina, Olga; Ahanotu, Onyemaechi; Sayegh, Allen; Weaver, James; Bechthold, Martin
year	2018
title	Fluid Equilibrium: Material Computation in Ferrofluidic Castings
doi	https://doi.org/10.52842/conf.acadia.2018.250
source	ACADIA // 2018: Recalibration. On imprecisionand infidelity. [Proceedings of the 38th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA) ISBN 978-0-692-17729-7] Mexico City, Mexico 18-20 October, 2018, pp. 250-259
summary	We present a computationally-based manufacturing process that allows for variable pattern casting through the use of ferrofluid – a mixture of suspended magnetic nanoparticles in a carrier liquid. The capacity of ferrofluid to form intricate spike and labyrinthine packing structures from ferrohydrodynamic instabilities is well recognized in industry and popular science. In this paper we employ these instabilities as a mold for the direct casting of rigid materials with complex periodic features. Furthermore, using a bitmap-based computational workflow and an array of high-strength neodymium magnets with linear staging, we demonstrate the ability to program the macro-scale pattern formation by modulating the magnetic field density within a single cast. Using this approach, it is possible to program specific patterns in the resulting cast tiles at both the micro- and macro-scale and thus generate tiled arrays with predictable halftone-like image features. We demonstrate the efficacy of this approach for a variety of materials typically used in the architecture, engineering, and construction industries (AEC) including epoxys, ceramics, and cements.
keywords	full paper, materials & adaptive systems, digital fabrication, digital materials, physics
series	ACADIA
type	paper
email
last changed	2022/06/07 08:00

For more results click below: