Defurnishing with X-Ray Vision: Joint Removal of Furniture from Panoramas and Mesh

TL;DR

This paper introduces a pipeline that combines mesh editing and panoramic inpainting to remove furniture from indoor scenes, producing clean, textured 3D models and panoramas with improved quality over existing methods.

Contribution

It proposes a novel joint approach using mesh segmentation, plane extension, hole filling, and ControlNet-guided inpainting to effectively defurnish indoor scenes in both mesh and panoramic formats.

Findings

Produces higher quality defurnished meshes and panoramas than neural radiance field methods.

Effectively removes furniture while preserving scene structure and details.

Outperforms RGB-D inpainting methods in visual quality and accuracy.

Abstract

We present a pipeline for generating defurnished replicas of indoor spaces represented as textured meshes and corresponding multi-view panoramic images. To achieve this, we first segment and remove furniture from the mesh representation, extend planes, and fill holes, obtaining a simplified defurnished mesh (SDM). This SDM acts as an ``X-ray'' of the scene's underlying structure, guiding the defurnishing process. We extract Canny edges from depth and normal images rendered from the SDM. We then use these as a guide to remove the furniture from panorama images via ControlNet inpainting. This control signal ensures the availability of global geometric information that may be hidden from a particular panoramic view by the furniture being removed. The inpainted panoramas are used to texture the mesh. We show that our approach produces higher quality assets than methods that rely on neural radiance fields, which tend to produce blurry low-resolution images, or RGB-D inpainting, which is highly susceptible to hallucinations.