Research map / 3D vision / NeRF and radiance fields
Ref-NPR: Reference-Based Non-Photorealistic Radiance Fields for Controllable Scene Stylization
Yuechen Zhang, Zexin He, Jinbo Xing, Xufeng Yao, Jiaya Jia
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 |
|---|---|---|---|
| Use reference-based ray registration and template-based semantic correspondence | Ref-NPR · Novel-view stylization | Ref-LPIPS: 0.394 -> 0.339 | Improved |
| Use reference-based ray registration and template-based semantic correspondence | Ref-NPR · Novel-view stylization | Robustness: 26.34 -> 28.11 | Improved |
| Use Ref-NPR rather than ARF | Stylization user study · Stylization sequences | Average rank: 2.52 -> 1.73 | Improved |
| Remove pseudo-ray supervision (reference loss) (ablation) | Ref-NPR · Stylized novel views | Visual quality | Hurt |
| Use reference-view rays only instead of registered pseudo-rays (ablation) | Ref-NPR · Stylized novel views | Texture detail | Hurt |
| Remove semantic feature supervision (ablation) | Ref-NPR · Occluded regions | Semantic stylization | Hurt |
| Remove coarse color-matching supervision (ablation) | Ref-NPR · Occluded regions | Color consistency | Hurt |
| Remove Template-based Correspondence Module (ablation) | Ref-NPR · Occluded regions | Semantic correspondence | Hurt |
| Remove smooth content update strategy (ablation) | Ref-NPR · Occluded regions | Stylization quality | Hurt |
| Use Texler rather than Ref-NPR | Novel-view stylization · Reference-related novel views | Ref-LPIPS: 0.339 -> 0.335 | Improved |
Later papers that compared with this method
6 later accepted papers report Ref-NPR as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.
- StylizedGS: Controllable Stylization for 3D Gaussian Splatting (TPAMI 2025 · better on 3 of 3 shared results)
- Multi-StyleGS: Stylized Gaussian Splatting with Multiple Styles (AAAI 2025 · better on 2 of 2 shared results)
- Geometry Transfer for Stylizing Radiance Fields (CVPR 2024 · better on 7 of 7 shared results)
- LAENeRF: Local Appearance Editing for Neural Radiance Fields (CVPR 2024 · better on 4 of 4 shared results)
- S-DyRF: Reference-Based Stylized Radiance Fields for Dynamic Scenes (CVPR 2024 · better on 2 of 4 shared results)
- ReGS: Reference-based Controllable Scene Stylization with Gaussian Splatting (NeurIPS 2024 · better on 3 of 3 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.
- SNeRF — S-NeRF: Neural Radiance Fields for Street Views (ICLR 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
- FPRF: Feed-Forward Photorealistic Style Transfer of Large-Scale 3D Neural Radiance Fields (AAAI 2024 · similarity 0.92)
- S-DyRF: Reference-Based Stylized Radiance Fields for Dynamic Scenes (CVPR 2024 · similarity 0.91)
- ReGS: Reference-based Controllable Scene Stylization with Gaussian Splatting (NeurIPS 2024 · similarity 0.90)
- Geometry Transfer for Stylizing Radiance Fields (CVPR 2024 · similarity 0.88)
- NeRF Analogies: Example-Based Visual Attribute Transfer for NeRFs (CVPR 2024 · similarity 0.87)
- Styl3R: Instant 3D Stylized Reconstruction for Arbitrary Scenes and Styles (NeurIPS 2025 · similarity 0.86)
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.
