Research map / 3D vision / NeRF and radiance fields
One Is All: Bridging the Gap between Neural Radiance Fields Architectures with Progressive Volume Distillation
Shuangkang Fang, Weixin Xu, Heng Wang, Yi Yang, Yufeng Wang, Shuchang Zhou
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 |
|---|---|---|---|
| Distill Hash teacher to MLP with PVD | NeRF on TanksAndTemples · TanksAndTemples | PSNR: 25.78 -> 27.50 | Improved |
| Distill Hash teacher to VM student with PVD | VM-decomposition on LLFF · LLFF | PSNR: 26.51 -> 25.73 | Hurt |
| Add volume-aligned feature loss (ablation) | PVD converting VM to MLP · NeRF-Synthetic | PSNR: 29.63 -> 30.49 | Improved |
| Add density loss (ablation) | PVD converting VM to MLP · NeRF-Synthetic | PSNR: 30.01 -> 30.49 | Improved |
| Add color loss (ablation) | PVD converting VM to MLP · NeRF-Synthetic | PSNR: 29.95 -> 30.49 | Improved |
| Add RGB loss (ablation) | PVD converting VM to MLP · NeRF-Synthetic | PSNR: 27.07 -> 30.49 | Improved |
| Add density-range constraint (ablation) | PVD converting VM to MLP · NeRF-Synthetic | PSNR: 28.45 -> 30.49 | Improved |
| Add block-wise distillation (ablation) | PVD converting VM to MLP · NeRF-Synthetic | PSNR: 29.62 -> 30.49 | Improved |
| Increase MLP split depth from K=2 to K=6 (ablation) | PVD Hash-to-MLP distillation · NeRF-Synthetic | PSNR: 30.39 -> 30.73 | Improved |
| Fine-tune a student after Hash-to-Tensors distillation | Sparse-tensor student · NeRF-Synthetic | PSNR: 27.85 -> 27.82 | No clear effect |
| Fine-tune a student after Tensors-to-Hash distillation | Hash student · NeRF-Synthetic | PSNR: 27.32 -> 32.41 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Plenoxels128 — Plenoxels: Radiance Fields Without Neural Networks (CVPR 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Progressive Volume Distillation with Active Learning for Efficient NeRF Architecture Conversion (IJCV 2026 · similarity 0.94)
- NeRFLight: Fast and Light Neural Radiance Fields Using a Shared Feature Grid (CVPR 2023 · similarity 0.93)
- Neural Radiance Fields From Sparse RGB-D Images for High-Quality View Synthesis (TPAMI 2022 · similarity 0.92)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.91)
- Mask-Based Modeling for Neural Radiance Fields (ICLR 2024 · similarity 0.91)
- Ced-NeRF: A Compact and Efficient Method for Dynamic Neural Radiance Fields (AAAI 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.
