Research map / 3D vision / NeRF and radiance fields
NeRFReN: Neural Radiance Fields With Reflections
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 |
|---|---|---|---|
| Replace a single NeRF with separate transmitted and reflected fields and image-domain composition | NeRF · RFFR scenes | PSNR: 36.48 -> 39.49 (art1); 32.22 -> 33.30 (mirror); 42.63 -> 42.33 (art2); 41.18 -> 41.03 (art3); 29.50 -> 30.18 (bookcase); | Mixed |
| Add depth smoothness prior (ablation) | NeRFReN · RFFR scenes | transmitted geometry and depth quality | Helped |
| Add bidirectional depth consistency prior (ablation) | NeRFReN · RFFR scenes | reflected-component cleanliness | Helped |
| Add both geometric priors (ablation) | NeRFReN · RFFR scenes | decomposition quality | Helped |
| Remove viewing-direction conditioning from transmitted radiance (ablation) | NeRFReN · bookcase | depth and decomposition quality | Hurt |
| Apply strong geometric constraints from the beginning instead of warming them up (ablation) | NeRFReN · RFFR scenes | decomposition quality | Hurt |
| Use viewing directions from the beginning instead of masking them early (ablation) | NeRFReN · RFFR scenes | reflection detail and decomposition quality | Hurt |
| Add reflection-mask supervision (ablation) | NeRFReN · mirror and tv | decomposition quality: 4 masks (mirror); 1 mask (tv) | Helped |
| Use NeRFReN rather than NeRF on scenes without strong reflections | NeRF · LLFF average | PSNR: 29.66 -> 29.16 | Hurt |
Closest papers by meaning
- Ref-NeRF: Structured View-Dependent Appearance for Neural Radiance Fields (CVPR 2022 · similarity 0.94)
- NeRFLight: Fast and Light Neural Radiance Fields Using a Shared Feature Grid (CVPR 2023 · similarity 0.92)
- D-NeRF: Neural Radiance Fields for Dynamic Scenes (CVPR 2021 · similarity 0.92)
- AligNeRF: High-Fidelity Neural Radiance Fields via Alignment-Aware Training (CVPR 2023 · similarity 0.92)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.91)
- NeRF-VPT: Learning Novel View Representations with Neural Radiance Fields via View Prompt Tuning (AAAI 2024 · similarity 0.91)
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.