Research map / 3D vision / Gaussian splatting
LumiMotion: Improving Gaussian Relighting with Scene Dynamics
Joanna Kaleta, Piotr Wójcik, Kacper Marzol, Tomasz Trzcinski, Kacper Kania, Marek Kowalski
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 Stage 2 inverse rendering (ablation) | LumiMotion · Golden Bay to Dam Wall | albedo PSNR: 23.935 -> 27.929 | Improved |
| add Stage 2 inverse rendering (ablation) | LumiMotion · Golden Bay to Dam Wall | relight LPIPS: 0.060 -> 0.048 | Improved |
| add temporal color change Delta c (ablation) | LumiMotion · Golden Bay to Dam Wall | albedo PSNR: 23.983 -> 27.929 | Improved |
| add temporal color change Delta c (ablation) | LumiMotion · Golden Bay to Dam Wall | albedo LPIPS: 0.083 -> 0.058 | Improved |
| add static-dynamic separation variable P (ablation) | LumiMotion · Golden Bay to Dam Wall | albedo PSNR: 26.768 -> 27.929 | Improved |
| add static-dynamic separation variable P (ablation) | LumiMotion · Golden Bay to Dam Wall | relight PSNR: 25.079 -> 25.405 | Improved |
| add static-dynamic separation variable P (ablation) | LumiMotion · Golden Bay to Dam Wall | relight SSIM: 0.933 -> 0.936 | Improved |
| increase separation weight or start separation earlier (ablation) | LumiMotion · dynamic scene separation parameters | shadow misclassification (qualitative) | Improved |
| train on dynamic scene with LumiMotion instead of static IR-GS | IR-GS · Dam Wall to Harbour Sunset | albedo PSNR: 22.888 -> 27.268 | Improved |
| train on dynamic scene with LumiMotion instead of static IR-GS | IR-GS · Golden Bay to Dam Wall | albedo LPIPS: 0.082 -> 0.058 | Improved |
| train on dynamic scene with LumiMotion instead of static IR-GS | IR-GS · Dam Wall to Harbour Sunset | relight PSNR: 26.177 -> 26.037 | No clear effect |
| train on dynamic scene with LumiMotion instead of static IR-GS | IR-GS · Dam Wall to Harbour Sunset | relight SSIM: 0.953 -> 0.928 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- IR-GS — IRGS: Inter-Reflective Gaussian Splatting with 2D Gaussian Ray Tracing (CVPR 2025) · better on 9 of 9 shared results
- GI-GS — GI-GS: Global Illumination Decomposition on Gaussian Splatting for Inverse Rendering (ICLR 2025) · better on 9 of 9 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
- GI-GS: Global Illumination Decomposition on Gaussian Splatting for Inverse Rendering (ICLR 2025 · similarity 0.91)
- GS-IR: 3D Gaussian Splatting for Inverse Rendering (CVPR 2024 · similarity 0.89)
- SVG-IR: Spatially-Varying Gaussian Splatting for Inverse Rendering (CVPR 2025 · similarity 0.89)
- MaterialRefGS: Reflective Gaussian Splatting with Multi-view Consistent Material Inference (NeurIPS 2025 · similarity 0.89)
- Subsurface Scattering for Gaussian Splatting (NeurIPS 2024 · similarity 0.89)
- Ref-GS: Directional Factorization for 2D Gaussian Splatting (CVPR 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.
