Research map / 3D vision / NeRF and radiance fields
Neural Fields for Structured Lighting
Aarrushi Shandilya, Benjamin Attal, Christian Richardt, James Tompkin, Matthew O'toole
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 structured light (dot pattern) supervision to NeRF (ablation) | NeRF · Synthetic scenes, 25 views | Depth MSE: 0.615 -> 0.013 | Improved |
| add structured light (dot pattern) supervision to NeRF (ablation) | NeRF · Synthetic scenes, 25 views | Normals MAE (degrees): 24.34 -> 2.84 | Improved |
| use dot pattern instead of flood illumination (ablation) | Ours · Synthetic scenes, 25 views | Normals MAE (degrees): 8.76 -> 2.84 | Improved |
| add structured light supervision (ablation) | NeRF · Synthetic scenes, 25 views | PSNR: 44.98 -> 44.13 | No clear effect |
| add structured light loss versus no depth supervision | NeRF · Real scenes, 2 views | PSNR: 27.53 -> 34.89 | Improved |
| add structured light loss versus dense RealSense depth supervision | NeRF · Real scenes, 2 views | PSNR: 36.61 -> 34.89 | Hurt |
| structured light loss versus dense depth supervision with many views | Ours · Real scenes, 4 views | PSNR: 40.74 -> 40.97 | No clear effect |
| add dense depth supervision to the method | Ours · Real scenes, 2 views | PSNR: 34.89 -> 36.70 | Improved |
| add structured light supervision to model scenes behind partially transparent mesh | Ours · Mesh and translucent box scenes | qualitative depth | Improved |
| explicitly predict normals with normal loss (ablation) | Ours · Lego and skateboard scenes | normal quality (qualitative) | Improved |
Closest papers by meaning
- Neural Relightable Participating Media Rendering (NeurIPS 2021 · similarity 0.89)
- ReNeRF: Relightable Neural Radiance Fields with Nearfield Lighting (ICCV 2023 · similarity 0.88)
- Light Field Neural Rendering (CVPR 2022 · similarity 0.88)
- Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering (NeurIPS 2021 · similarity 0.88)
- A Pre-convolved Representation for Plug-and-Play Neural Illumination Fields (AAAI 2024 · similarity 0.87)
- TurboSL: Dense Accurate and Fast 3D by Neural Inverse Structured Light (CVPR 2024 · similarity 0.87)
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.
