360-DFPE: Leveraging Monocular 360-Layouts for Direct Floor Plan Estimation

TL;DR

This paper introduces 360-DFPE, a novel method for direct floor plan estimation from monocular 360-images, combining visual SLAM and layout estimation to sequentially generate accurate floor plans without active sensors.

Contribution

The paper presents a new sequential floor plan estimation approach that integrates monocular visual SLAM with 360-room layout estimation, including novel algorithms for scale alignment, room identification, and shape refinement.

Findings

Achieves competitive accuracy compared to sensor-based methods

Handles large-scale scenes with challenging layouts

Provides a new dataset with point clouds and sequential images

Abstract

We present 360-DFPE, a sequential floor plan estimation method that directly takes 360-images as input without relying on active sensors or 3D information. Our approach leverages a loosely coupled integration between a monocular visual SLAM solution and a monocular 360-room layout approach, which estimate camera poses and layout geometries, respectively. Since our task is to sequentially capture the floor plan using monocular images, the entire scene structure, room instances, and room shapes are unknown. To tackle these challenges, we first handle the scale difference between visual odometry and layout geometry via formulating an entropy minimization process, which enables us to directly align 360-layouts without knowing the entire scene in advance. Second, to sequentially identify individual rooms, we propose a novel room identification algorithm that tracks every room along the camera exploration using geometry information. Lastly, to estimate the final shape of the room, we propose a shortest path algorithm with an iterative coarse-to-fine strategy, which improves prior formulations with higher accuracy and faster run-time. Moreover, we collect a new floor plan dataset with challenging large-scale scenes, providing both point clouds and sequential 360-image information. Experimental results show that our monocular solution achieves favorable performance against the current state-of-the-art algorithms that rely on active sensors and require the entire scene reconstruction data in advance.