Research map / 3D vision / NeRF and radiance fields
Nerfies: Deformable Neural Radiance Fields
Keunhong Park, Utkarsh Sinha, Jonathan T. Barron, Sofien Bouaziz, Dan B Goldman, Steven M. Seitz, Ricardo Martin-Brualla
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 elastic regularization (ablation) | Nerfies · quasi-static scenes | LPIPS: .300 -> .282 | Improved |
| add elastic regularization (ablation) | Nerfies · dynamic scenes | LPIPS: .170 -> .168 | No clear effect |
| increase elastic weight lambda from 0.001 to 0.01 (ablation) | Nerfies · dynamic scenes | LPIPS: .168 -> .156 | Improved |
| increase elastic weight lambda from 0.001 to 0.01 (ablation) | Nerfies · quasi-static scenes | LPIPS: .282 -> .281 | No clear effect |
| add coarse-to-fine regularization (ablation) | Nerfies · dynamic scenes | LPIPS: .228 -> .168 | Improved |
| add coarse-to-fine regularization (ablation) | Nerfies · quasi-static scenes | LPIPS: .301 -> .282 | Improved |
| use SE(3) deformation field instead of translation field (ablation) | Nerfies · dynamic scenes | LPIPS: .177 -> .168 | Improved |
| use SE(3) deformation field instead of translation field (ablation) | Nerfies · quasi-static scenes | LPIPS: .293 -> .282 | Improved |
| add background regularization (ablation) | Nerfies · quasi-static scenes | PSNR: 21.5 -> 23.7 | Improved |
| add background regularization (ablation) | Nerfies · dynamic scenes | PSNR: 21.8 -> 22.1 | Improved |
| add all contributions (SE3, elastic, BG, coarse-to-fine) (ablation) | Ours (base) · dynamic scenes | LPIPS: .275 -> .168 | Improved |
| add per-image latent code to NeRF | NeRF · quasi-static scenes | LPIPS: .502 -> .452 | Improved |
Closest papers by meaning
- D-NeRF: Neural Radiance Fields for Dynamic Scenes (CVPR 2021 · similarity 0.92)
- NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collections (CVPR 2021 · similarity 0.91)
- UP-NeRF: Unconstrained Pose Prior-Free Neural Radiance Field (NeurIPS 2023 · similarity 0.90)
- NeRF-Editing: Geometry Editing of Neural Radiance Fields (CVPR 2022 · similarity 0.90)
- Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Video (ICCV 2021 · similarity 0.90)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.89)
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.
