Research map / 3D vision / NeRF and radiance fields
Neural Sparse Voxel Fields
Read the paperCheck 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 NeRF with NSVF | NeRF · Synthetic-NeRF | PSNR: 31.01 -> 31.74 | Helped |
| Replace NeRF with NSVF | NeRF · Synthetic-NSVF | PSNR: 30.81 -> 35.13 | Helped |
| Replace NeRF with NSVF | NeRF · BlendedMVS | PSNR: 24.15 -> 26.90 | Helped |
| Replace NeRF with NSVF | NeRF · Tanks and Temples | PSNR: 25.78 -> 28.40 | Helped |
| Replace NeRF with NSVF | NeRF · Multi-scene Synthetic-NeRF + Synthetic-NSVF | PSNR: 25.71 -> 30.68 | Helped |
| Add early termination with ε=0.010 (ablation) | NSVF · Wineholder | Speed: 4.0 -> 2.0 | Helped |
| Add early termination with ε=0.010 (ablation) | NSVF · Wineholder | PSNR: 31.93 -> 32.04 | No clear effect |
| Raise the early termination threshold to ε=0.100 (ablation) | NSVF · Wineholder | PSNR: 32.04 -> 29.99 | Hurt |
| Increase progressive training rounds from 1 to 4 (ablation) | NSVF · Wineholder | PSNR: 28.82 -> 30.89 | Helped |
| Reduce voxel size from 1 to 1/8 (ablation) | NSVF · Wineholder | PSNR: 28.82 -> 30.17 | Helped |
| Add the depth loss (ablation) | NSVF · ScanNet (one scene) | RMSE: 0.210 -> 0.079 | Helped |
| Add the depth loss (ablation) | NSVF · ScanNet (one scene) | PSNR: 25.07 -> 25.48 | Helped |
Closest papers by meaning
- Sparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering (CVPR 2025 · similarity 0.92)
- CVT-xRF: Contrastive In-Voxel Transformer for 3D Consistent Radiance Fields from Sparse Inputs (CVPR 2024 · similarity 0.88)
- Learning with Unreliability: Fast Few-shot Voxel Radiance Fields with Relative Geometric Consistency (CVPR 2024 · similarity 0.88)
- DynaVol: Unsupervised Learning for Dynamic Scenes through Object-Centric Voxelization (ICLR 2024 · similarity 0.87)
- SCube: Instant Large-Scale Scene Reconstruction using VoxSplats (NeurIPS 2024 · similarity 0.87)
- Zero-Shot Novel View and Depth Synthesis with Multi-View Geometric Diffusion (CVPR 2025 · 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.