---
_id: '14070'
abstract:
- lang: eng
  text: "Deploying digital twins across the broader building stock is constrained
    by a key bottleneck: high-quality 3D content from LiDAR or photogrammetry remains
    labor-intensive and difficult to scale. We introduce Generative Twins (GT), a
    workflow that shifts digital twin creation from explicit geometric reconstruction
    toward probabilistic 3D synthesis. The pipeline takes a\r\nsingle street-level
    façade photograph, corrects perspective distortion and removes occlusions, uses
    an instruction-conditioned multimodal image model to synthesise an isometric view
    of the building, and passes that view to an image-to-3D reconstruction model that
    leverages learned shape priors to produce a textured mesh without manual modelling.
    Three image to-3D platforms (Hitem3D, Hyper3D and Hunyuan3D 3.1) are compared
    on identical isometric inputs and assessed on architectural boundary clarity.
    We demonstrate GT through two applied cases: (i) reconstruction of ordinary residential
    streets in Detmold as stimuli for a virtual-reality study of multi sensory urban
    perception, and (ii) Green Editor, a design-exploration tool in which façade greenery
    is added or removed at the image layer before mesh generation.\r\nTogether these
    cases position Generative Twins as a scalable pathway for the digitization of
    ordinary building stock, with emphasis on controllability, editability, and iterative
    design workflows."
author:
- first_name: Mohamed Khaled
  full_name: Ashmawy, Mohamed Khaled
  id: '81286'
  last_name: Ashmawy
- first_name: Buse
  full_name: Akay, Buse
  id: '88405'
  last_name: Akay
- first_name: Alvaro
  full_name: Balderrama, Alvaro
  id: '74171'
  last_name: Balderrama
  orcid: 0000-0002-4118-2838
citation:
  ama: 'Ashmawy MK, Akay B, Balderrama A. <i>Towards Generative Twins: An AI Based
    Pipeline from Explicit 3D Reconstruction to Probabilistic Synthesis</i>. Vol 1.
    eCAADe; 2026. doi:<a href="https://doi.org/10.52842/conf.ecaade.2026.1.619">10.52842/conf.ecaade.2026.1.619</a>'
  apa: 'Ashmawy, M. K., Akay, B., &#38; Balderrama, A. (2026). Towards Generative
    Twins: An AI Based Pipeline from Explicit 3D Reconstruction to Probabilistic Synthesis.
    In <i>eCAADe proceedings</i> (Vol. 1). eCAADe. <a href="https://doi.org/10.52842/conf.ecaade.2026.1.619">https://doi.org/10.52842/conf.ecaade.2026.1.619</a>'
  bjps: '<b>Ashmawy MK, Akay B and Balderrama A</b> (2026) <i>Towards Generative Twins:
    An AI Based Pipeline from Explicit 3D Reconstruction to Probabilistic Synthesis</i>.
    eCAADe.'
  chicago: 'Ashmawy, Mohamed Khaled, Buse Akay, and Alvaro Balderrama. <i>Towards
    Generative Twins: An AI Based Pipeline from Explicit 3D Reconstruction to Probabilistic
    Synthesis</i>. <i>ECAADe Proceedings</i>. Vol. 1. eCAADe, 2026. <a href="https://doi.org/10.52842/conf.ecaade.2026.1.619">https://doi.org/10.52842/conf.ecaade.2026.1.619</a>.'
  chicago-de: 'Ashmawy, Mohamed Khaled, Buse Akay und Alvaro Balderrama. 2026. <i>Towards
    Generative Twins: An AI Based Pipeline from Explicit 3D Reconstruction to Probabilistic
    Synthesis</i>. <i>eCAADe proceedings</i>. Bd. 1. eCAADe. doi:<a href="https://doi.org/10.52842/conf.ecaade.2026.1.619">10.52842/conf.ecaade.2026.1.619</a>,
    .'
  din1505-2-1: '<span style="font-variant:small-caps;">Ashmawy, Mohamed Khaled</span>
    ; <span style="font-variant:small-caps;">Akay, Buse</span> ; <span style="font-variant:small-caps;">Balderrama,
    Alvaro</span>: <i>Towards Generative Twins: An AI Based Pipeline from Explicit
    3D Reconstruction to Probabilistic Synthesis</i>. Bd. 1 : eCAADe, 2026'
  havard: 'M.K. Ashmawy, B. Akay, A. Balderrama, Towards Generative Twins: An AI Based
    Pipeline from Explicit 3D Reconstruction to Probabilistic Synthesis, eCAADe, 2026.'
  ieee: 'M. K. Ashmawy, B. Akay, and A. Balderrama, <i>Towards Generative Twins: An
    AI Based Pipeline from Explicit 3D Reconstruction to Probabilistic Synthesis</i>,
    vol. 1. eCAADe, 2026. doi: <a href="https://doi.org/10.52842/conf.ecaade.2026.1.619">10.52842/conf.ecaade.2026.1.619</a>.'
  mla: 'Ashmawy, Mohamed Khaled, et al. “Towards Generative Twins: An AI Based Pipeline
    from Explicit 3D Reconstruction to Probabilistic Synthesis.” <i>ECAADe Proceedings</i>,
    vol. 1, eCAADe, 2026, <a href="https://doi.org/10.52842/conf.ecaade.2026.1.619">https://doi.org/10.52842/conf.ecaade.2026.1.619</a>.'
  short: 'M.K. Ashmawy, B. Akay, A. Balderrama, Towards Generative Twins: An AI Based
    Pipeline from Explicit 3D Reconstruction to Probabilistic Synthesis, eCAADe, 2026.'
  ufg: '<b>Ashmawy, Mohamed Khaled/Akay, Buse/Balderrama, Alvaro</b>: Towards Generative
    Twins: An AI Based Pipeline from Explicit 3D Reconstruction to Probabilistic Synthesis,
    Bd. 1, o. O. 2026.'
  van: 'Ashmawy MK, Akay B, Balderrama A. Towards Generative Twins: An AI Based Pipeline
    from Explicit 3D Reconstruction to Probabilistic Synthesis. Vol. 1, eCAADe proceedings.
    eCAADe; 2026.'
conference:
  end_date: 2026-09-11
  location: Lübeck
  name: 44. eCAADe-Konferenz
  start_date: 2026-09-07
date_created: 2026-09-30T10:14:15Z
date_updated: 2026-10-01T14:28:23Z
department:
- _id: DEP1634
doi: 10.52842/conf.ecaade.2026.1.619
intvolume: '         1'
keyword:
- Digital Twins
- Virtual Reality (VR
- Artificial Intelligence (AI)
- Image-to-3D
- Generative Design.
language:
- iso: eng
publication: eCAADe proceedings
publication_identifier:
  issn:
  - 2684-1843
publication_status: published
publisher: eCAADe
status: public
title: 'Towards Generative Twins: An AI Based Pipeline from Explicit 3D Reconstruction
  to Probabilistic Synthesis'
type: conference_editor_article
user_id: '83781'
volume: 1
year: '2026'
...