Research map / 3D vision / NeRF and radiance fields
Dynamic PlenOctree for Adaptive Sampling Refinement in Explicit NeRF
Haotian Bai, Yiqi Lin, Yize Chen, Lin Wang
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 |
|---|---|---|---|
| replace fixed octree with dynamic octree calibration (DOT) | PlenOctree · NeRF-Synthetic (RTX3090) | FPS: 250 -> 452 | Improved |
| replace fixed octree with dynamic octree calibration (DOT) | PlenOctree · Tanks & Temples | memory (GB): 2.62 -> 0.82 | Improved |
| prune with hierarchical feature fusion instead of heuristic rejection (ablation) | DOT prune-only vs POT · NeRF-Synthetic | PSNR: over 60% memory reduction with competitive or better PSNR | Improved |
| add complementary sampling to pruning (ablation) | DOT · NeRF-Synthetic | PSNR: +0.15 dB at about 10% memory cost | Improved |
| use Q (ray weight) instead of sigma as pruning target (ablation) | DOT · NeRF-Synthetic | FPS: about 20% FPS gain, similar quality | Improved |
| add recursive pruning (ablation) | DOT · NeRF-Synthetic | PSNR / memory: 5-8% more memory reduction, about 0.1 PSNR lower | Mixed |
| raise pruning threshold tau from 1 to 10 (ablation) | DOT · NeRF-Synthetic | PSNR / memory: about 3% of POT's parameters; quality bottleneck on complex scenes | Mixed |
| apply DOT compression (median-cut quantization of SH) | DOT on MX150 2GB GPU · NeRF-Synthetic (MX150) | FPS: PlenOctree unavailable -> 11 FPS (full), 12 FPS (recursive) | Improved |
Closest papers by meaning
- PlenOctrees for Real-Time Rendering of Neural Radiance Fields (ICCV 2021 · similarity 0.93)
- Fourier PlenOctrees for Dynamic Radiance Field Rendering in Real-Time (CVPR 2022 · similarity 0.90)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.89)
- Ced-NeRF: A Compact and Efficient Method for Dynamic Neural Radiance Fields (AAAI 2024 · similarity 0.87)
- NeRF-VPT: Learning Novel View Representations with Neural Radiance Fields via View Prompt Tuning (AAAI 2024 · similarity 0.87)
- A General Implicit Framework for Fast NeRF Composition and Rendering (AAAI 2024 · similarity 0.87)
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.
