Research map / 3D vision / Face and shape
Neural Lumigraph Rendering
Petr Kellnhofer, Lars C. Jebe, Andrew Jones, Ryan Spicer, Kari Pulli, Gordon Wetzstein
Check 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 an SDF-based SIREN scene representation instead of a standard ReLU MLP (ablation) | Neural Lumigraph Rendering · Ablation scene, detailed image representation | image reconstruction quality | Improved |
| Add angular smoothness loss to the SIREN radiance field (ablation) | Neural Lumigraph Rendering · Ablation scene, view interpolation | interpolated-view quality | Improved |
| Increase the batch size for training (ablation) | Neural Lumigraph Rendering · Ablation scene | spatial image quality | Improved |
| Add Fourier features for ray direction (ablation) | Neural Lumigraph Rendering · Ablation scene | radiance-field quality | Improved |
| Increase neural texture view density from 1× to 3× (ablation) | NLR-RAS · A1 novel interpolated/extrapolated view | PSNR: 23.83 -> 30.33 | Improved |
| Increase neural texture view density from 1× to 3× (ablation) | NLR-RAS · Average across seven head scenes | PSNR: 23.11 -> 28.89 | Improved |
| Use L1 rather than L2 image reconstruction loss (ablation) | Neural Lumigraph Rendering · Training image reconstruction | image sharpness | Improved |
| Increase angular smoothness weight to large values (ablation) | Neural Lumigraph Rendering · Training | shape quality | Hurt |
| Use the NLR surface with neural textures instead of IDR geometry and textures (ablation) | Real-time rasterized rendering · Head scenes, novel interpolated/extrapolated views | PSNR | Improved |
| Use NLR neural textures instead of IDR neural textures (ablation) | Real-time rasterized rendering · Head scenes, novel views | PSNR | Improved |
Closest papers by meaning
- NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis (ECCV 2020 · similarity 0.88)
- Fast Training of Neural Lumigraph Representations using Meta Learning (NeurIPS 2021 · similarity 0.87)
- Neural 3D Video Synthesis From Multi-View Video (CVPR 2022 · similarity 0.87)
- UNIS: A Unified Framework for Achieving Unbiased Neural Implicit Surfaces in Volume Rendering (ICCV 2025 · similarity 0.87)
- HybridNeRF: Efficient Neural Rendering via Adaptive Volumetric Surfaces (CVPR 2024 · similarity 0.87)
- What You See is What You GAN: Rendering Every Pixel for High-Fidelity Geometry in 3D GANs (CVPR 2024 · 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.
