Metraipolis › archive › dc_a1881bb58d4b

Depth-to-RGB: Repurposing a Frozen Depth Estimator for Geometry-Guided Compositing — arXiv:2610.09125

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09125
License (as archived)
CC0-1.0
Archived
2026-10-08T12:46:57.235Z
SHA-256 of the archived text
71ed1a86fd616e7efdcf37d067c17b014e053b3b7e46440d875f89325716534a
Stable URL of this page
https://metraipolis.com/documento/dc_a1881bb58d4b

How to cite

Depth-to-RGB: Repurposing a Frozen Depth Estimator for Geometry-Guided Compositing — arXiv:2610.09125 — arXiv (en), https://arxiv.org/abs/2610.09125 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a1881bb58d4b (Metraipolis archive dc_a1881bb58d4b, sha256 71ed1a86fd616e7efdcf37d067c17b014e053b3b7e46440d875f89325716534a)
BibTeX
@misc{dca1881bb58d4b,
  title = {Depth-to-RGB: Repurposing a Frozen Depth Estimator for Geometry-Guided Compositing — arXiv:2610.09125},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09125},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a1881bb58d4b; sha256 71ed1a86fd616e7efdcf37d067c17b014e053b3b7e46440d875f89325716534a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09125. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reference-based object compositing inserts or replaces an object using a background image, a reference image, and a 2D compositing mask. These inputs guide appearance and placement but leave the completed scene's geometry implicit, which can distort object structure or alter the surroundings. Our Depth-to-RGB (D2R) framework predicts composite depth for a scene not yet observed in the RGB inputs. It learns reference-conditioned corrections to a frozen depth estimator using encoder features of paired completed scenes as targets. The unchanged decoder maps the corrected representation to the intended scene's depth, which a separately trained renderer holds fixed during RGB synthesis. Under matched architecture and training, encoder-feature supervision reduces OOD Stage-1 AbsRel by 31.4% relative to decoded-depth supervision. We also introduce AnyInsertion++ with paired in-distribution and category-disjoint splits to evaluate generalization beyond compositing training categories. The complete D2R system leads 12 open-source and 3 closed-source baselines in estimator-derived geometry and photometric quality on both paired splits. On category-disjoint data, D2R reduces AbsRel by 43.7% and improves PSNR by 2.4 dB over the matched RGB baseline. Across three unpaired benchmarks, D2R leads both identity metrics and reduces mean CLIP reference cosine distance by 55% relative to the strongest baseline. Project page: this https URL

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.09125 — 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 2026-10-08T12:18:15.311Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).