A Two-Stage Diffusion Framework for 3D Building Reconstruction from Single Nadir Images

dc.contributor.advisorSohn, Gunho
dc.contributor.authorKorny, Youssef Shaaban Mohammed
dc.date.accessioned2026-11-06T16:16:28Z
dc.date.available2026-11-06T16:16:28Z
dc.date.copyright2026-06-11
dc.date.issued2026-11-06
dc.date.updated2026-11-06T16:16:27Z
dc.degree.disciplineEarth & Space Science
dc.degree.levelMaster's
dc.degree.nameMSc - Master of Science
dc.description.abstractReconstructing 3D urban geometry from single nadir aerial images is an ill-posed problem due to severe vertical ambiguity and occluded building facades. To resolve this, this thesis proposes GeoPriorPC, a novel two-stage generative framework bridging foundational 2D image priors and unstructured 3D point clouds. Stage 1 fine-tunes a Latent Diffusion Model (LDM) to extract a compressed geometric prior, specifically, oblique surface normals, from flat nadir inputs. Stage 2 employs a Diffusion Point Transformer (DiPT) backbone. It utilizes a dual-conditioning mechanism, integrating global adaptive Layer Normalization (adaLN-Zero) and local spatial cross-attention, to structurally guide the 3D denoising process using the learned 2D prior. Evaluated on the AIM2PC benchmark, GeoPriorPC establishes a new state-of-the-art, achieving a 44% relative improvement in F1-score and reducing Earth Mover's Distance by 49%. By explicitly resolving depth ambiguity, this framework successfully synthesizes the accurate architectural manifolds critical for urban digital twins.
dc.identifier.urihttps://hdl.handle.net/10315/44336
dc.languageen
dc.rightsAuthor owns copyright, except where explicitly noted. Please contact the author directly with licensing requests.
dc.subjectArtificial intelligence
dc.subjectRemote sensing
dc.subject.keywords3D building reconstruction
dc.subject.keywordsGenerative diffusion models
dc.subject.keywordsPoint cloud generation
dc.subject.keywordsNadir aerial imagery
dc.subject.keywordsRemote sensing
dc.subject.keywordsLatent Diffusion Models (LDM)
dc.subject.keywordsGeometric priors
dc.subject.keywordsSurface normal estimation
dc.subject.keywordsDeep learning
dc.subject.keywordsComputer vision
dc.titleA Two-Stage Diffusion Framework for 3D Building Reconstruction from Single Nadir Images
dc.typeElectronic Thesis or Dissertation

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Korny_Youssef_Shaaban_Mohammed_2026_MSc.pdf
Size:
36.22 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 2 of 2
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.87 KB
Format:
Plain Text
Description:
Loading...
Thumbnail Image
Name:
YorkU_ETDlicense.txt
Size:
3.39 KB
Format:
Plain Text
Description: