Saved in:
Bibliographic Details
Main Authors: Slavcheva, Mira, Gausebeck, Dave, Chen, Kevin, Buchhofer, David, Sabik, Azwad, Ma, Chen, Dhillon, Sachal, Brandt, Olaf, Dolhasz, Alan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.03682
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • We propose a pipeline that leverages Stable Diffusion to improve inpainting results in the context of defurnishing -- the removal of furniture items from indoor panorama images. Specifically, we illustrate how increased context, domain-specific model fine-tuning, and improved image blending can produce high-fidelity inpaints that are geometrically plausible without needing to rely on room layout estimation. We demonstrate qualitative and quantitative improvements over other furniture removal techniques.