Metraipolis › archive › dc_83f9e91cde1c
ClearGS: Reliability-Aware Gaussian Splatting from Handheld Videos — arXiv:2609.31509
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.31509
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T01:39:59.774Z
- SHA-256 of the archived text
5f6de5332d3e7b123af8a040fc69eebb2f79890eb2bbc93d0765e20d5e56e3c7- Stable URL of this page
https://metraipolis.com/documento/dc_83f9e91cde1c
How to cite
ClearGS: Reliability-Aware Gaussian Splatting from Handheld Videos — arXiv:2609.31509 — arXiv (en), https://arxiv.org/abs/2609.31509 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_83f9e91cde1c (Metraipolis archive dc_83f9e91cde1c, sha256 5f6de5332d3e7b123af8a040fc69eebb2f79890eb2bbc93d0765e20d5e56e3c7)
BibTeX
@misc{dc83f9e91cde1c,
title = {ClearGS: Reliability-Aware Gaussian Splatting from Handheld Videos — arXiv:2609.31509},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.31509},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_83f9e91cde1c; sha256 5f6de5332d3e7b123af8a040fc69eebb2f79890eb2bbc93d0765e20d5e56e3c7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31509. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We present ClearGS for 3D Gaussian Splatting (3DGS) from handheld videos with uneven viewpoint coverage and mixed frame quality. Rather than selecting frames with binary decisions, ClearGS uses Reliability-aware View Allocation (RVA) to assign graded raw-supervision weights based on appearance reliability, degradation risk, and geometric utility, while weakly reactivating useful suppressed frames to maintain trajectory coverage. Since weighting cannot restore details lost to blur or distortion, ClearGS further introduces Render-Guided In-Video Restoration (RIVR). The current 3DGS render provides a pose-aligned structural candidate, a frozen no-reference restoration expert restores the corresponding raw video observation without any clean reference image, and no-reference perceptual scores select among the render, restored observation, and high-frequency fused candidate. ClearGS then applies Full-Trajectory Repair Consolidation to revisit accepted repairs and preserve details introduced early. On GS2E and GSOTM, ClearGS achieves state-of-the-art overall performance, with consistent CLIP-IQA and MUSIQ gains and LPIPS reductions in most degradation settings, without paired sharp supervision or matched clean references. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.31509 — 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 2026-10-08T01:12:14.610Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).