Research map / 3D vision / NeRF and radiance fields
Dynamic Neural Radiance Field From Defocused Monocular Video
Xianrui Luo, Huiqiang Sun, Juewen Peng, Zhiguo Cao
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 |
|---|---|---|---|
| model defocus blur with layered DoF volume rendering inside a dynamic NeRF | D2RF · Synthesized defocused dynamic scenes (8 VDW scenes, all) | LPIPS: 0.209 -> 0.130 (vs NSFF) | Improved |
| model blur in 3D with D2RF instead of 2D deblurring preprocessing plus dynamic NeRF | RoDynRF / DVS · Synthesized defocused dynamic scenes (8 VDW scenes, all) | PSNR: 25.79 -> 27.30 (vs 2D deblur + RoDynRF) | Improved |
| add 2D single-image deblurring preprocessing before dynamic NeRF | RoDynRF · Synthesized defocused dynamic scenes (8 VDW scenes, all) | PSNR: 26.18 -> 25.79 PSNR; 0.227 -> 0.196 LPIPS | No clear effect |
| add 2D deblurring preprocessing before DVS | DVS · Synthesized defocused dynamic scenes (8 VDW scenes, all) | PSNR: 25.43 -> 24.52 | Hurt |
| add cross-time rendering (ablation) | D2RF · Synthesized defocused dynamic scenes (8 VDW scenes, whole image) | PSNR: 22.61 -> 27.30 | Improved |
| add layered DoF volume rendering (ablation) | D2RF · Synthesized defocused dynamic scenes (8 VDW scenes, whole image) | LPIPS: 0.211 -> 0.130 | Improved |
| use optimized sparse kernel instead of full regular kernel (ablation) | D2RF · Synthesized defocused dynamic scenes (8 VDW scenes, dynamic regions) | LPIPS: 0.246 -> 0.162 | Improved |
| add separate static scene representation (ablation) | D2RF · Synthesized defocused dynamic scenes (8 VDW scenes, whole image) | PSNR: 26.20 -> 27.30 | Improved |
| layered DoF volume rendering and optimized kernel on PSNR in whole image (ablation) | D2RF · Synthesized defocused dynamic scenes (8 VDW scenes, whole image) | PSNR: 27.11 -> 27.30 (layered volume); 27.25 -> 27.30 (optimized kernel) | No clear effect |
| model defocus in volume rendering | D2RF · Training memory | GPU memory (G): 11.9 -> 20.1 (vs NSFF) | Hurt |
Closest papers by meaning
- DyBluRF: Dynamic Neural Radiance Fields from Blurry Monocular Video (CVPR 2024 · similarity 0.93)
- Deblur-NeRF: Neural Radiance Fields From Blurry Images (CVPR 2022 · similarity 0.92)
- DynaMoDe-NeRF: Motion-aware Deblurring Neural Radiance Field for Dynamic Scenes (CVPR 2025 · similarity 0.91)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.89)
- Mitigating Motion Blur in Neural Radiance Fields with Events and Frames (CVPR 2024 · similarity 0.89)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.88)
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.
