Academic Journal
From NURBS to neural networks: Efficient geometry encoding for generative AI in architectural design.
| Τίτλος: | From NURBS to neural networks: Efficient geometry encoding for generative AI in architectural design. |
|---|---|
| Συγγραφείς: | Sebestyen, Adam, Wiltsche, Albert, Stavric, Milena, Özdenizci, Ozan |
| Πηγή: | International Journal of Architectural Computing; Sep2025, Vol. 23 Issue 3, p720-741, 22p |
| Θεματικοί όροι: | Generative artificial intelligence, Architectural design, Data reduction, Parametric equations, Computer-aided design software, Artificial neural networks |
| Περίληψη: | The existing 3D representations for AI models, such as meshes, voxels, signed distance functions, and point clouds, are not compatible with architectural design workflows that rely on NURBS geometry, which is mainly used in CAD programs. These formats lead to large datasets, high computational costs, and loss of geometric precision, limiting further usability in CAD software. This research introduces a novel methodology for encoding NURBS geometries into compact, tensor-based NumPy data for training generative AI models and vice versa. Our methodology involves the design of comparative experiments, the comparison of NURBS tensor representations with other 3D representations, and the use of reconstruction accuracy as a key metric to evaluate performance. Custom components for the Rhinoceros 3D parametric environment Grasshopper were developed enabling bidirectional conversion between NURBS geometry and NumPy tensors. These components are being released as a Grasshopper plugin under the name Wiener Dog as a free download. Our approach maintains geometric accuracy, reduces data size, and integrates seamlessly with existing deep learning libraries. The proposed methodology was tested on datasets of helicoid surfaces and lofted polysurfaces, demonstrating high reconstruction accuracy and generative potential. The ultimate aim is to build an AI tool that aids in exploring the great variety of geometric forms for architectural design. [ABSTRACT FROM AUTHOR] |
| Copyright of International Journal of Architectural Computing is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Βάση Δεδομένων: | Complementary Index |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: edb DbLabel: Complementary Index An: 187842912 RelevancyScore: 1023 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 1023.08752441406 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: From NURBS to neural networks: Efficient geometry encoding for generative AI in architectural design. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Sebestyen%2C+Adam%22">Sebestyen, Adam</searchLink><br /><searchLink fieldCode="AR" term="%22Wiltsche%2C+Albert%22">Wiltsche, Albert</searchLink><br /><searchLink fieldCode="AR" term="%22Stavric%2C+Milena%22">Stavric, Milena</searchLink><br /><searchLink fieldCode="AR" term="%22Özdenizci%2C+Ozan%22">Özdenizci, Ozan</searchLink> – Name: TitleSource Label: Source Group: Src Data: International Journal of Architectural Computing; Sep2025, Vol. 23 Issue 3, p720-741, 22p – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Generative+artificial+intelligence%22">Generative artificial intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Architectural+design%22">Architectural design</searchLink><br /><searchLink fieldCode="DE" term="%22Data+reduction%22">Data reduction</searchLink><br /><searchLink fieldCode="DE" term="%22Parametric+equations%22">Parametric equations</searchLink><br /><searchLink fieldCode="DE" term="%22Computer-aided+design+software%22">Computer-aided design software</searchLink><br /><searchLink fieldCode="DE" term="%22Artificial+neural+networks%22">Artificial neural networks</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: The existing 3D representations for AI models, such as meshes, voxels, signed distance functions, and point clouds, are not compatible with architectural design workflows that rely on NURBS geometry, which is mainly used in CAD programs. These formats lead to large datasets, high computational costs, and loss of geometric precision, limiting further usability in CAD software. This research introduces a novel methodology for encoding NURBS geometries into compact, tensor-based NumPy data for training generative AI models and vice versa. Our methodology involves the design of comparative experiments, the comparison of NURBS tensor representations with other 3D representations, and the use of reconstruction accuracy as a key metric to evaluate performance. Custom components for the Rhinoceros 3D parametric environment Grasshopper were developed enabling bidirectional conversion between NURBS geometry and NumPy tensors. These components are being released as a Grasshopper plugin under the name Wiener Dog as a free download. Our approach maintains geometric accuracy, reduces data size, and integrates seamlessly with existing deep learning libraries. The proposed methodology was tested on datasets of helicoid surfaces and lofted polysurfaces, demonstrating high reconstruction accuracy and generative potential. The ultimate aim is to build an AI tool that aids in exploring the great variety of geometric forms for architectural design. [ABSTRACT FROM AUTHOR] – Name: Abstract Label: Group: Ab Data: <i>Copyright of International Journal of Architectural Computing is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edb&AN=187842912 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1177/14780771251353791 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 22 StartPage: 720 Subjects: – SubjectFull: Generative artificial intelligence Type: general – SubjectFull: Architectural design Type: general – SubjectFull: Data reduction Type: general – SubjectFull: Parametric equations Type: general – SubjectFull: Computer-aided design software Type: general – SubjectFull: Artificial neural networks Type: general Titles: – TitleFull: From NURBS to neural networks: Efficient geometry encoding for generative AI in architectural design. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Sebestyen, Adam – PersonEntity: Name: NameFull: Wiltsche, Albert – PersonEntity: Name: NameFull: Stavric, Milena – PersonEntity: Name: NameFull: Özdenizci, Ozan IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 09 Text: Sep2025 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 14780771 Numbering: – Type: volume Value: 23 – Type: issue Value: 3 Titles: – TitleFull: International Journal of Architectural Computing Type: main |
| ResultId | 1 |