Metraipolis › archive › dc_a46c5cece715

Guiding Image-to-3D Generation with Test-Time Partial Observations — arXiv:2609.10531

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.10531
License (as archived)
CC0-1.0
Archived
2026-10-16T11:00:08.823Z
SHA-256 of the archived text
5ca185c8d4b3b917f7e29fba83f0edb5b0f59fd1f7ee8949ee5d8cbcde41ff1a
Stable URL of this page
https://metraipolis.com/documento/dc_a46c5cece715

How to cite

Guiding Image-to-3D Generation with Test-Time Partial Observations — arXiv:2609.10531 — arXiv (en), https://arxiv.org/abs/2609.10531 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a46c5cece715 (Metraipolis archive dc_a46c5cece715, sha256 5ca185c8d4b3b917f7e29fba83f0edb5b0f59fd1f7ee8949ee5d8cbcde41ff1a)
BibTeX
@misc{dca46c5cece715,
  title = {Guiding Image-to-3D Generation with Test-Time Partial Observations — arXiv:2609.10531},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.10531},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a46c5cece715; sha256 5ca185c8d4b3b917f7e29fba83f0edb5b0f59fd1f7ee8949ee5d8cbcde41ff1a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10531. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Image-to-3D models can generate visually compelling 3D assets from a single RGB image, but their geometry is often only loosely constrained by the available observations, limiting their use in applications that require geometric fidelity. In many real-world settings, however, partial geometric observations of the object may be available at test time. We introduce a training-free framework for incorporating such evidence into pretrained image-to-3D generative models without retraining or finetuning. To do this, we guide generation using a ray-consistent observation likelihood defined over the model's occupancy representation, combining surface occupancy and free-space evidence. Applied to SAM 3D and its multi-view extension, our approach substantially improves geometric fidelity across different levels of observability, as well as visual quality. Our results demonstrate that pretrained image-to-3D models can effectively integrate partial geometric observations through explicit test-time guidance, complementing their learned generative priors without modifying the underlying model.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-10 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.10531 — 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 2026-10-16T10:02:34.412Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).