Floorplan Restoration by Structure Hallucinating Transformer Cascades

TL;DR

This paper introduces a new neural network architecture that reconstructs complete floorplans from partial data by hallucinating unseen structures, significantly advancing the state-of-the-art in architectural reconstruction.

Contribution

It proposes a novel cascade Transformer-based neural network for extreme floorplan reconstruction, including a new benchmark dataset and evaluation methods.

Findings

Outperforms existing techniques on a dataset of 701 houses

Effectively reconstructs invisible architectural structures

Demonstrates superior qualitative and quantitative results

Abstract

This paper presents an extreme floorplan reconstruction task, a new benchmark for the task, and a neural architecture as a solution. Given a partial floorplan reconstruction inferred or curated from panorama images, the task is to reconstruct a complete floorplan including invisible architectural structures. The proposed neural network 1) encodes an input partial floorplan into a set of latent vectors by convolutional neural networks and a Transformer; and 2) reconstructs an entire floorplan while hallucinating invisible rooms and doors by cascading Transformer decoders. Qualitative and quantitative evaluations demonstrate effectiveness of our approach over the benchmark of 701 houses, outperforming the state-of-the-art reconstruction techniques. We will share our code, models, and data.