Floor Plan Reconstruction from Sparse Views: Combining Graph Neural Network with Constrained Diffusion - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Floor Plan Reconstruction from Sparse Views: Combining Graph Neural Network with Constrained Diffusion

Résumé

We address the challenging problem of floor plan reconstruction from sparse views and a room-connectivity graph. As a first stage, we construct a flexible graph-structure unifying the connectivity graph and the sparse observed data. Using our Graph Neural Network architecture, we can then refine the available information and predict unobserved room properties. In a second step, we introduce a Constrained Diffusion Model to reconstruct consistent floor plan matching the available information, despite of its sparsity. More precisely, we use a Cross-Attention mechanism armed with shape descriptors to guarantee that the generated floor plan reflects both the input room connectivity and the geometry observed in the sparse views.
Fichier principal
Vignette du fichier
gueze_iccv_cvaad_2023.pdf (591.55 Ko) Télécharger le fichier
thumbnail.jpg (52.49 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04216274 , version 1 (24-09-2023)

Identifiants

  • HAL Id : hal-04216274 , version 1

Citer

Arnaud Gueze, Matthieu Ospici, Damien Rohmer, Marie-Paule Cani. Floor Plan Reconstruction from Sparse Views: Combining Graph Neural Network with Constrained Diffusion. ICCV Workshop on Computer Vision Aided Architectural Design, Oct 2023, Paris, France. ⟨hal-04216274⟩
87 Consultations
237 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More