Research map / 3D vision / Gaussian splatting
3D Gaussian Texture for Real-time Mesoscale Appearance Synthesis and Rendering
Xiang Chen, Jia Li, Lu Wang, Beibei Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add opacity-aware blending (ablation) | Gaussian texture · Mesoscale synthesis | SIFID: 2.27 -> 1.43 | Improved |
| add multi-layer texture (ablation) | Gaussian texture · Mesoscale synthesis | SIFID: 1.69 -> 1.43 | Improved |
| add variance-preserving synthesis to autocovariance synthesis (ablation) | Gaussian texture · Brick synthesis | SIFID: 1.35 -> 0.87 | Improved |
| use autocovariance-preserving instead of random synthesis (ablation) | Gaussian texture · Brick synthesis | SIFID: 1.81 -> 1.35 | Improved |
| add scaling constraint loss to mapping (ablation) | Gaussian texture mapping · Mapping to objects | visual artifacts | Improved |
| add encoder regularizing latent space during training (ablation) | Gaussian texture · Synthesis on plane | visual artifacts | Improved |
| use Gaussian texture instead of NeRF-Texture | Ours · Grass | SIFID: 15.94 -> 10.86 | Improved |
| use Gaussian texture instead of NeRF-Texture | Ours · Durian | SIFID: 10.71 -> 4.82 | Improved |
| use Gaussian texture instead of NeRF-Texture | Ours · Wicker | SIFID: 6.89 -> 4.38 | Improved |
| use Gaussian texture instead of NeRF-Texture | Ours · Flower | SIFID: 6.96 -> 5.26 | Improved |
| use Gaussian texture instead of NeRF-Texture | Ours · Grass | memory: 1.1 GB -> 280 MB | Improved |
| use Gaussian texture instead of NeRF-Texture | Ours · Synthesis and rendering | speed: about 500x acceleration | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- NeRF-Texture — NeRF-Texture: Synthesizing Neural Radiance Field Textures (TPAMI 2024) · better on 8 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Textured Gaussians for Enhanced 3D Scene Appearance Modeling (CVPR 2025 · similarity 0.92)
- IBGS: Image-Based Gaussian Splatting (NeurIPS 2025 · similarity 0.91)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.89)
- Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesis (NeurIPS 2024 · similarity 0.88)
- Spacetime Gaussian Feature Splatting for Real-Time Dynamic View Synthesis (CVPR 2024 · similarity 0.88)
- Decoupling Appearance Variations with 3D Consistent Features in Gaussian Splatting (AAAI 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
