arXiv：cs.AI· De Jiang, Peiqiang Wang, Kehong Yuan, Shaohua Ma·· 4 小时前AI 评分37
AdaTex4D：为 4D 高斯泼溅自适应分配纹理容量
ADATEX4D: adaptive texture capacity allocation for 4D gaussian splatting
AI 导读
AdaTex4D 是一个面向基于形变的 4D Gaussian Splatting 的自适应纹理容量模块，让每个 Gaussian 携带打包 RGBA triplanes，其两轴依据可见性归一化的屏幕空间梯度和形变局部尺度独立生长。
正文
Abstract:Textured Gaussians improve local appearance capacity, but assigning the same texture resolution to every primitive wastes storage on low-detail or weakly visible regions. We introduce AdaTex4D, an adaptive texture-capacity module for deformation-based 4D Gaussian Splatting. Each Gaussian carries packed RGBA triplanes whose two axes grow independently according to visibility normalized screen-space gradients and deformed local scales. Experiments on N3DV and PanopticSports show that AdaTex4D reduces texture storage by more than half while preserving reconstruction quality. Under fixed memory budgets, adaptive allocation also improves quality over uniform texture assignment and reduces overall model and peak memory. These results show that dynamic, anisotropic texture allocation provides a more efficient way to distribute local appearance capacity in 4D Gaussian representations.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2609.29963 [cs.CV] |
| (or arXiv:2609.29963v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2609.29963 arXiv-issued DOI via DataCite |
Submission history
From: De Jiang [view email]
[v1]
Thu, 24 Sep 2026 15:19:23 UTC (2,669 KB)
[v2]
Fri, 2 Oct 2026 15:46:21 UTC (1,495 KB)
来源：arXiv：cs.AI · arxiv.org