Research map / 3D vision / Gaussian splatting
StylizedGS: Controllable Stylization for 3D Gaussian Splatting
Dingxi Zhang, Yu-Jie Yuan, Zhuoxun Chen, Fang‐Lue Zhang, Zhenliang He, Shiguang Shan, Lin Gao
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 color transfer (recoloring) (ablation) | StylizedGS · Ablation (50 stylized cases) | ArtFID: 31.57 -> 28.29 | Improved |
| fine-tune density (opacity) along with color (ablation) | StylizedGS · Ablation (50 stylized cases) | ArtFID: 38.54 -> 28.29 | Improved |
| fine-tune density (opacity) along with color (ablation) | StylizedGS · Ablation (50 stylized cases) | SSIM: 0.53 -> 0.55 | Improved |
| add depth preservation loss (ablation) | StylizedGS · Ablation (50 stylized cases) | SSIM: 0.35 -> 0.55 | Improved |
| add depth preservation loss (ablation) | StylizedGS · Ablation (50 stylized cases) | ArtFID: 31.20 -> 28.29 | Improved |
| add filter-based Gaussian refinement (ablation) | StylizedGS · Ablation (50 stylized cases) | ArtFID: 30.02 -> 28.29 | Improved |
| add filter-based Gaussian refinement (ablation) | StylizedGS · Ablation (50 stylized cases) | SSIM: 0.39 -> 0.55 | Improved |
| use conv2 and conv3 VGG blocks for NNFM loss (ablation) | StylizedGS · Feature layer choice (qualitative) | style detail | Improved |
| StylizedGS vs ARF training time | StylizedGS · Stylization novel views (LLFF/T&T) | training time: 2.25 min -> 0.87 min | Improved |
| StylizedGS vs StyleGaussian rendering speed | StylizedGS · Stylization novel views (LLFF/T&T) | FPS: 142 -> 153 | Improved |
| StylizedGS vs ARF long-range consistency | StylizedGS · Long-range consistency | LPIPS: 0.185 -> 0.112 | Improved |
Later papers that compared with this method
2 later accepted papers report StylizedGS as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- Geometry-Aware Style Transfer in 3D Gaussian Splatting (ECCV 2026 · better on 1 of 5 shared results)
- Capacity-Controlled Multi-View Stylization of 3D Gaussian Splatting (ECCV 2026 · better on 1 of 2 shared results)
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.
- Ref-NPR — Ref-NPR: Reference-Based Non-Photorealistic Radiance Fields for Controllable Scene Stylization (CVPR 2023) · better on 3 of 3 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
- Multi-StyleGS: Stylized Gaussian Splatting with Multiple Styles (AAAI 2025 · similarity 0.94)
- ReGS: Reference-based Controllable Scene Stylization with Gaussian Splatting (NeurIPS 2024 · similarity 0.93)
- Morpheus: Text-Driven 3D Gaussian Splat Shape and Color Stylization (CVPR 2025 · similarity 0.92)
- Styl3R: Instant 3D Stylized Reconstruction for Arbitrary Scenes and Styles (NeurIPS 2025 · similarity 0.92)
- SGSST: Scaling Gaussian Splatting Style Transfer (CVPR 2025 · similarity 0.90)
- CLIPGaussian: Universal and Multimodal Style Transfer Based on Gaussian Splatting (NeurIPS 2025 · similarity 0.89)
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.
