EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ResEdit: Residual embeddings for precise generative image editing

Canberk Baykal, Valentin Deschaintre, Yannick Hold‐Geoffroy, Michael Fischer, Anna Frühstück, Cengiz Öztireli, Iliyan Georgiev

Computer Graphics Forum · 20.08.2026

DOI: 10.1111/cgf.70551 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Graphics Forum, online erschienen am 20.08.2026. Verfasst von Canberk Baykal und 6 weiteren Autor:innen. Die Arbeit nennt 77 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Conditional diffusion image generators can be repurposed for editing through inversion, without the need for large‐scale paired fine‐tuning data. However, producing high‐quality, targeted edits while maintaining image identity and global consistency remains challenging, as weakly conditioned inversion often embeds conflicting image features into the noise. We demonstrate that incorporating a residual image encoding as additional conditioning enables both improved identity preservation and better editability. We optimize this residual encoding to provide a strong conditioning signal for reconstruction, thereby reducing the reliance on inversion and susceptibility to its aforementioned pitfalls. To ensure this residual does not interfere with desired edits, we incorporate a gradient reversal‐based optimization strategy that disentangles the residual from the edited condition. We illustrate our method's ability to produce high‐fidelity results across precise intrinsic‐based editing and relighting, and show proof‐of‐concept text‐guided manipulation. Project page: johnberg1.github.io/resedit

Zitieren

APA

Baykal, C., Deschaintre, V., Hold‐Geoffroy, Y., Fischer, M., Frühstück, A., Öztireli, C., & Georgiev, I. (2026). ResEdit: Residual embeddings for precise generative image editing. Computer Graphics Forum. https://doi.org/10.1111/cgf.70551

Vancouver

Baykal C, Deschaintre V, Hold‐Geoffroy Y, Fischer M, Frühstück A, Öztireli C, et al. ResEdit: Residual embeddings for precise generative image editing. Computer Graphics Forum. 2026. doi:10.1111/cgf.70551

BibTeX
@article{Baykal2026,
  author = {Baykal, Canberk and Deschaintre, Valentin and Hold‐Geoffroy, Yannick and Fischer, Michael and Frühstück, Anna and Öztireli, Cengiz and Georgiev, Iliyan},
  title = {ResEdit: Residual embeddings for precise generative image editing},
  journal = {Computer Graphics Forum},
  year = {2026},
  doi = {10.1111/cgf.70551}
}
Kostenlos starten →