Abstract:We introduce a hybrid Gaussian-hash-grid radiance representation for reconstructing 2D Gaussian scene models from multi-view images. Similar to NeST splatting, our approach reduces the entanglement between geometry and appearance common in NeRF-based models, but adds per-Gaussian latent features alongside hash-grid features to bias the optimizer toward a separation of low- and high-frequency scene components. This explicit frequency-based decomposition reduces the tendency of high-frequency texture to compensate for geometric errors. Encouraging Gaussians with hard opacity falloffs further strengthens the separation between geometry and appearance, improving both geometry reconstruction and rendering efficiency. Finally, probabilistic pruning combined with a sparsity-inducing BCE opacity loss allows redundant Gaussians to be turned off, yielding a minimal set of Gaussians sufficient to represent the scene. Using both synthetic and real-world datasets, we compare against the state of the art in Gaussian-based novel-view synthesis and demonstrate superior reconstruction fidelity with an order of magnitude fewer primitives.
| Comments: | 22 pages, 9 figures |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR) |
| Cite as: | arXiv:2604.14928 [cs.CV] |
| (or arXiv:2604.14928v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2604.14928 arXiv-issued DOI via DataCite |
From: Neel Kelkar [view email]
[v1]
Thu, 16 Apr 2026 12:13:09 UTC (30,196 KB)
[v2]
Fri, 17 Apr 2026 16:12:24 UTC (30,195 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。