Metraipolis › archive › dc_8fd0240fcd7f
Filling the Unseen: Scene Extrapolation via 3D Gaussian Splatting — arXiv:2609.13262
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.13262
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T01:52:34.409Z
- SHA-256 of the archived text
23fed57d7f88dbc26e04a3b1855655a067cafa0a6db1d9e16492ff5b5e1e46da- Stable URL of this page
https://metraipolis.com/documento/dc_8fd0240fcd7f
How to cite
Filling the Unseen: Scene Extrapolation via 3D Gaussian Splatting — arXiv:2609.13262 — arXiv (en), https://arxiv.org/abs/2609.13262 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8fd0240fcd7f (Metraipolis archive dc_8fd0240fcd7f, sha256 23fed57d7f88dbc26e04a3b1855655a067cafa0a6db1d9e16492ff5b5e1e46da)
BibTeX
@misc{dc8fd0240fcd7f,
title = {Filling the Unseen: Scene Extrapolation via 3D Gaussian Splatting — arXiv:2609.13262},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.13262},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8fd0240fcd7f; sha256 23fed57d7f88dbc26e04a3b1855655a067cafa0a6db1d9e16492ff5b5e1e46da},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.13262. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
3D Gaussian Splatting achieves photorealistic reconstruction within training view distribution, yet it degrades on out-of-distribution novel views, exhibiting holes in unobserved regions and artifacts in observable areas. Recent works formulate this task as extrapolation and interpolation and try to address it with generative models, but remain limited in extrapolation scale and quality. They repeat a generate-reconstruct-shift cycle to progressively build a scene, which introduces accumulated errors with every step conditioning on previous outcomes. In this work, we propose a holistic framework for extrapolation and interpolation. We devise an independent camera view detection mechanism to enable parallel conflict-free extrapolation, circumventing the reliance on the aforementioned error-prone cycle. Building upon this, we design a hierarchical pipeline that extrapolates independent and dependent camera views separately. Additionally, previous methods overlook inconsistency between generated and original images, resulting in compromising well-reconstructed areas. We propose a plug-and-play Quality-Aware Mask (QA-Mask) module, enabling selective utilization of generated data. By calibrating learning weights with pixel-wise rendering quality, it prevents generation-induced degradations on well-constructed areas. Extensive experiments demonstrate the superior performance of our framework, with QA-Mask generalizing to multiple generative reconstruction models. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.13262 — 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 2026-10-16T01:25:23.481Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).