Metraipolis › archive › dc_0a35c2571dfc

Compact Feed-Forward 3D Gaussians via Saliency-Guided Primitive Merging — arXiv:2608.10712

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.10712
License (as archived)
CC0-1.0
Archived
2026-10-19T06:30:25.320Z
SHA-256 of the archived text
4b6ccd331a7eea1a5c0b279f3547ad8c75534f6fee48afff618fc68f49bcffb6
Stable URL of this page
https://metraipolis.com/documento/dc_0a35c2571dfc

How to cite

Compact Feed-Forward 3D Gaussians via Saliency-Guided Primitive Merging — arXiv:2608.10712 — arXiv (en), https://arxiv.org/abs/2608.10712 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0a35c2571dfc (Metraipolis archive dc_0a35c2571dfc, sha256 4b6ccd331a7eea1a5c0b279f3547ad8c75534f6fee48afff618fc68f49bcffb6)
BibTeX
@misc{dc0a35c2571dfc,
  title = {Compact Feed-Forward 3D Gaussians via Saliency-Guided Primitive Merging — arXiv:2608.10712},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.10712},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0a35c2571dfc; sha256 4b6ccd331a7eea1a5c0b279f3547ad8c75534f6fee48afff618fc68f49bcffb6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10712. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

3D scene reconstruction, modeling, and rendering are highly relevant for numerous tasks, and 3D Gaussian splatting has become a standard choice in this context. Its feed-forward variants provide fast reconstruction from sparse input views but often produce per-pixel primitives, leading to highly redundant and thus inefficient representations. We present a structure-aware merging pipeline that takes per-pixel primitives from any feed-forward method and consolidates them into a compact, content-adaptive Gaussian set while largely retaining visual quality at just $\frac{1}{20}^\text{th}$ of the Gaussians of a per-pixel method. We group spatially coherent Gaussians of similar appearance into variable-size clusters via adaptive superpixel segmentation guided by a saliency map, which allocates fine segments to textured regions and coarse segments to homogeneous areas. We compress each cluster into a compact latent representation through a learned encoder, then match and consolidate representations across views based on geometric overlap and feature similarity via a learned merger. A level-of-detail decoder then produces the final Gaussians at a controllable resolution, enabling a flexible quality-efficiency trade-off at inference. As a post-processing module, the pipeline is backbone-agnostic, leveraging the strengths of existing feed-forward methods. This leads to better and more robust quality than achieved by previous approaches that target a reduction in primitive count, while providing a highly compact representation, that can be rendered efficiently.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-12 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.10712 — 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 2026-10-19T06:25:27.276Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).