Metraipolis › archive › dc_a582f3032dfd

Training-Free Inpainting Across Domains with a Frozen Text-to-Image Diffusion Model — arXiv:2609.00862

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.00862
License (as archived): CC0-1.0
Archived: 2026-10-17T07:13:40.072Z
SHA-256 of the archived text: cb6d309589baf60d9d97f47e74f8ffbbef8b0b91877989e4202b3ebc88fa89a9
Stable URL of this page: https://metraipolis.com/documento/dc_a582f3032dfd

How to cite

Training-Free Inpainting Across Domains with a Frozen Text-to-Image Diffusion Model — arXiv:2609.00862 — arXiv (en), https://arxiv.org/abs/2609.00862 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a582f3032dfd (Metraipolis archive dc_a582f3032dfd, sha256 cb6d309589baf60d9d97f47e74f8ffbbef8b0b91877989e4202b3ebc88fa89a9)

BibTeX

@misc{dca582f3032dfd,
  title = {Training-Free Inpainting Across Domains with a Frozen Text-to-Image Diffusion Model — arXiv:2609.00862},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.00862},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a582f3032dfd; sha256 cb6d309589baf60d9d97f47e74f8ffbbef8b0b91877989e4202b3ebc88fa89a9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00862. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We show that a frozen generic text-to-image diffusion model can perform conditional inpainting across three evaluated natural-image domains with one fixed controller configuration, without inpainting-specific weight training, dataset-specific weight adaptation, or learned inpainting-specific conditioning channels. Step-PI augments known-region projection with boundary-interior latent feedback, persistent PI state, and a predefined four-field release schedule that modulates controller signals along the reverse trajectory. Developed only on Main35-disjoint CelebA-HQ pilots, the controller transfers unchanged to AFHQ and Places2. Across two field-identical comparisons on the same 3,500 cases, adding persistent state and replacing uniform release with the predefined schedule each improve all 15 dataset-metric cells; 95% bootstrap intervals exclude zero for all five metrics in both comparisons. In descriptive native-route comparisons, Step-PI leads LanPaint and PILOT (the closest evaluated training-free baselines using vanilla SD1.5) on all five equal-dataset macro metrics. Inpainting-trained systems retain the absolute metric leads but rely on substantial inpainting-specific offline optimization. Our method provides a complementary approach for repurposing a frozen generic text-to-image model for cross-domain inpainting through test-time latent control.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.00862 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T07:00:58.940Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).