Research map / 3D vision / Depth and stereo
Lighthouse: Predicting Lighting Volumes for Spatially-Coherent Illumination
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 a completed multiscale 3D lighting volume instead of the strongest baseline's independent lighting-map predictions | Lighthouse vs. Neural Illumination · InteriorNet test, all content | PSNR (dB): 16.59 ± 1.91 -> 17.29 ± 1.97 | Helped |
| Use a completed multiscale 3D lighting volume instead of the strongest baseline's independent lighting-map predictions | Lighthouse vs. Neural Illumination · InteriorNet test, all content | Angular Error (°): 5.26 ± 2.84 -> 4.71 ± 2.68 | Helped |
| Use a completed multiscale 3D lighting volume instead of the strongest baseline's independent lighting-map predictions | Lighthouse vs. Neural Illumination · InteriorNet test, observed content | PSNR (dB): 18.58 ± 3.55 -> 19.79 ± 3.99 | Helped |
| Use a completed multiscale 3D lighting volume instead of the strongest baseline's independent lighting-map predictions | Lighthouse vs. Neural Illumination · InteriorNet test, observed content | Angular Error (°): 4.97 ± 3.42 -> 3.76 ± 3.09 | Helped |
| Add multiscale volume completion to the MPI-only representation (ablation) | Lighthouse · InteriorNet test, all content | PSNR (dB): 15.26 ± 2.11 -> 17.29 ± 1.97 | Helped |
| Add multiscale volume completion to the MPI-only representation (ablation) | Lighthouse · InteriorNet test, all content | Angular Error (°): 6.41 ± 3.42 -> 4.71 ± 2.68 | Helped |
| Add multiscale volume completion to the MPI-only representation (ablation) | Lighthouse · InteriorNet test, observed content | PSNR (dB): 17.97 ± 3.86 -> 19.79 ± 3.99 | Helped |
| Add multiscale volume completion to the MPI-only representation (ablation) | Lighthouse · InteriorNet test, observed content | Angular Error (°): 4.45 ± 4.46 -> 3.76 ± 3.09 | Helped |
| Add the adversarial loss (ablation) | Lighthouse · InteriorNet test, all content | PSNR (dB): 17.54 ± 1.97 -> 17.29 ± 1.97 | Hurt |
| Add the adversarial loss (ablation) | Lighthouse · InteriorNet test, all content | Angular Error (°): 4.74 ± 2.70 -> 4.71 ± 2.68 | Helped |
| Add the adversarial loss (ablation) | Lighthouse · InteriorNet test, observed content | PSNR (dB): 19.74 ± 3.64 -> 19.79 ± 3.99 | No clear effect |
| Add the adversarial loss (ablation) | Lighthouse · InteriorNet test, observed content | Angular Error (°): 4.18 ± 3.29 -> 3.76 ± 3.09 | Helped |
Closest papers by meaning
- MetaGS: A Meta-Learned Gaussian-Phong Model for Out-of-Distribution 3D Scene Relighting (NeurIPS 2025 · similarity 0.87)
- ROGR: Relightable 3D Objects using Generative Relighting (NeurIPS 2025 · similarity 0.87)
- ReCap: Better Gaussian Relighting with Cross-Environment Captures (CVPR 2025 · similarity 0.87)
- LightOctree: Lightweight 3D Spatially-Coherent Indoor Lighting Estimation (CVPR 2024 · similarity 0.87)
- LuxDiT: Lighting Estimation with Video Diffusion Transformer (NeurIPS 2025 · similarity 0.86)
- UniLumos: Fast and Unified Image and Video Relighting with Physics-Plausible Feedback (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.