Research map / 3D vision / NeRF and radiance fields
Learning Neural Light Fields With Ray-Space Embedding
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 |
|---|---|---|---|
| add local affine ray-space embedding instead of no embedding (ablation) | Neural light field with subdivision · Real Forward-Facing | PSNR: 26.696 -> 27.454 | Helped |
| add feature-space embedding instead of no embedding (ablation) | Neural light field with subdivision · Real Forward-Facing | PSNR: 26.696 -> 26.922 | Helped |
| replace feature-space embedding with local affine embedding (ablation) | Neural light field without subdivision · Stanford Light Field | PSNR: 37.120 -> 38.054 | Helped |
| add local affine ray-space embedding instead of no embedding (ablation) | Neural light field without subdivision · Stanford Light Field | PSNR: 25.111 -> 38.054 | Helped |
| increase subdivision resolution from 4³ to 32³ (ablation) | Subdivided neural light field · Real Forward-Facing | PSNR: 25.579 -> 27.454 | Helped |
| increase subdivision resolution from 8³ to 32³ (ablation) | Subdivided neural light field · Real Forward-Facing | FPS: 1.42 -> 0.34 | Hurt |
| increase subdivision resolution from 8³ to 32³ (ablation) | Subdivided neural light field · Real Forward-Facing | PSNR: 26.860 -> 27.454 | Helped |
| move the initial ray-parameterization plane from z = 3 to z = 0 (ablation) | Neural light field without embedding · Amethyst | PSNR: 21.491 -> 29.631 | Helped |
| use neural light field with subdivision instead of NeRF | Real Forward-Facing · Real Forward-Facing | FPS: 0.07 -> 0.34 | Helped |
| use neural light field with subdivision instead of NeRF | Real Forward-Facing · Real Forward-Facing | PSNR: 27.928 -> 27.454 | Mixed |
| use neural light field with subdivision instead of NeX | Shiny · Shiny | PSNR: 28.268 -> 28.499 | Helped |
| use neural light field without subdivision instead of NeRF | Stanford Light Field · Stanford Light Field | PSNR: 37.559 -> 38.054 | Helped |
Closest papers by meaning
- Light Field Neural Rendering (CVPR 2022 · similarity 0.93)
- Neural Point Light Fields (CVPR 2022 · similarity 0.92)
- Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering (NeurIPS 2021 · similarity 0.92)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.91)
- A Pre-convolved Representation for Plug-and-Play Neural Illumination Fields (AAAI 2024 · similarity 0.90)
- Learning Robust Generalizable Radiance Field with Visibility and Feature Augmented Point Representation (ICLR 2024 · similarity 0.90)
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.