Metraipolis › archive › dc_3ac2ef853709

Laplacian Frequency Hierarchies for Efficient 3D Gaussian Splatting Training — arXiv:2609.03334

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.03334
License (as archived)
CC0-1.0
Archived
2026-10-17T02:56:13.599Z
SHA-256 of the archived text
ab8b9d0a2b36b4f9942996b6d489e4c152d4e242aff38a916502d1cc3be1be42
Stable URL of this page
https://metraipolis.com/documento/dc_3ac2ef853709

How to cite

Laplacian Frequency Hierarchies for Efficient 3D Gaussian Splatting Training — arXiv:2609.03334 — arXiv (en), https://arxiv.org/abs/2609.03334 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3ac2ef853709 (Metraipolis archive dc_3ac2ef853709, sha256 ab8b9d0a2b36b4f9942996b6d489e4c152d4e242aff38a916502d1cc3be1be42)
BibTeX
@misc{dc3ac2ef853709,
  title = {Laplacian Frequency Hierarchies for Efficient 3D Gaussian Splatting Training — arXiv:2609.03334},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.03334},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3ac2ef853709; sha256 ab8b9d0a2b36b4f9942996b6d489e4c152d4e242aff38a916502d1cc3be1be42},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03334. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A key bottleneck in 3D Gaussian Splatting training is the continual growth of Gaussian primitives, which increases optimization cost and slows convergence, especially at high resolutions. We propose Laplacian Frequency Hierarchies, a simple yet efficient 3DGS scheme that combines Laplacian image decomposition with coarse-to-fine, frequency-staged training. After fitting lower-frequency structure, we archive the corresponding Gaussian field so that subsequent fields can optimize higher-frequency residuals without carrying the full primitive burden, and we compose the rendered components in the image domain via a Laplacian-style reconstruction at inference time. This design reduces the number of active Gaussians during training, thereby lowering optimization overhead and accelerating training. The proposed scheme is plug-and-play and orthogonal to prior 3DGS accelerations: it can be directly combined with strong backbones such as Taming-3DGS and FastGS to improve training speed with competitive reconstruction quality. It achieves average speedups of 1.73x and 1.21x at 1K setting, and 1.74x and 1.33x at 4K setting on Taming-3DGS and FastGS, with larger gains on more challenging scenes and increasingly pronounced benefits at higher resolutions.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-04 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.03334 — 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 2026-10-17T02:27:18.947Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).