An Empty Room Is All We Want: Automatic Defurnishing of Indoor Panoramas

Mira Slavcheva, Dave Gausebeck, Kevin Chen, David Buchhofer, Azwad Sabik, Chen Ma, Sachal Dhillon, Olaf Brandt, Alan Dolhasz

CVPRW 2024 pp. 7384-7394

doi:10.1109/CVPRW63382.2024.00734 /cvprw/2024/slavcheva2024cvprw-empty/

Abstract

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.

PDF CVPRW Semantic Scholar

Cite

Text

Slavcheva et al. "An Empty Room Is All We Want: Automatic Defurnishing of Indoor Panoramas." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2024. doi:10.1109/CVPRW63382.2024.00734

Markdown

[Slavcheva et al. "An Empty Room Is All We Want: Automatic Defurnishing of Indoor Panoramas." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2024.](https://mlanthology.org/cvprw/2024/slavcheva2024cvprw-empty/) doi:10.1109/CVPRW63382.2024.00734

BibTeX

@inproceedings{slavcheva2024cvprw-empty,
  title     = {{An Empty Room Is All We Want: Automatic Defurnishing of Indoor Panoramas}},
  author    = {Slavcheva, Mira and Gausebeck, Dave and Chen, Kevin and Buchhofer, David and Sabik, Azwad and Ma, Chen and Dhillon, Sachal and Brandt, Olaf and Dolhasz, Alan},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2024},
  pages     = {7384-7394},
  doi       = {10.1109/CVPRW63382.2024.00734},
  url       = {https://mlanthology.org/cvprw/2024/slavcheva2024cvprw-empty/}
}