LiteTex-GS speeds up and simplifies texturing for 3D view synthesis
LiteTex-GS: Fast and Lightweight Texturing for Gaussian Splatting
Computer Vision and Pattern Recognition
Summary
Creating new views of 3D scenes on a computer can be slow and memory-heavy when textures are detailed. The authors introduce LiteTex-GS, a method that adds only small amounts of texture where needed and removes unimportant parts to keep things simple. This approach helps keep the process fast and uses less memory without losing image quality. Their experiments show it works well compared to older methods that are slower and larger.
What this means in practice
- •For virtual reality developers: Generate high-quality virtual scenes faster with less memory by using efficient texturing methods for Gaussian-based 3D models.
- •For game graphics programmers: Create realistic textures on 3D objects in real-time while reducing computational load during game rendering.
Authors
Zhiwei Li, Yijia Guo, Yishi Lu, Liwen Hu, Hong Rao, Shengbo Chen, Lei Ma
Abstract
Gaussian Splatting has enabled real-time novel view synthesis, but its tightly coupled geometry and appearance representation often require a large number of primitives to reproduce high-frequency texture details, leading to substantial memory and optimization costs. Recent textured 2D Gaussian methods alleviate this limitation by attaching texture maps to Gaussian primitives. However, bridging the fundamental structural gap between discrete Gaussians and continuous 2D grids requires complex parameterizations that introduce severe computational overhead. This overhead fundamentally compromises the original efficiency of Gaussian Splatting, making the balance between detailed texturing and computational agility an unresolved challenge. To address these challenges, we propose LiteTex-GS, a fast and lightweight texturing framework for Gaussian Splatting. Our method initializes an extremely compact representation, assigning minimal local texture to each Gaussian and progressively allocates higher resolution only to primitives with significant reconstruction errors. To maintain a streamlined geometric scaffold, we introduce a contribution- and area-aware pruning strategy that eliminates low-utility Gaussians. Furthermore, to mitigate the gradient dilution caused by texture upsampling, we design a resolution-aware update rule that preserves rapid and stable convergence. Extensive experiments on standard novel view synthesis benchmarks demonstrate that our method achieves competitive or superior rendering quality while using substantially fewer parameters and less training time than existing textured Gaussian baselines.