Research map / 3D vision / NeRF and radiance fields
StructNeRF: Neural Radiance Fields for Indoor Scenes With Structural Hints
Zheng Chen, Chen Wang, Yuan-Chen Guo, Song–Hai Zhang
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 dense sampling of patch-match points instead of keypoints (ablation) | StructNeRF · NYUv2 | PSNR: 27.18 -> 28.10 | Improved |
| add patch-based instead of point-based photometric loss (ablation) | StructNeRF · NYUv2 | PSNR: 28.03 -> 28.10 | Improved |
| add warm-up training for sparse depth loss (ablation) | StructNeRF · NYUv2 | PSNR: 27.60 -> 28.10 | Improved |
| add sparse COLMAP depth priors (ablation) | StructNeRF · NYUv2 | Depth RMSE: 0.4052 -> 0.3113 | Improved |
| add patch-match multi-view photometric loss (ablation) | StructNeRF · NYUv2 | PSNR: 27.56 -> 28.10 | Improved |
| add plane regularization (ablation) | StructNeRF · NYUv2 | PSNR: 27.92 -> 28.10 | Improved |
| add plane regularization (ablation) | StructNeRF · NYUv2 | Plane Mean Dev: 0.0321 -> 0.0301 | Mixed |
| add dense sampling (ablation) | StructNeRF · NYUv2 | Depth RMSE: 0.3246 -> 0.3113 | Improved |
| add per-frame latent code (test code zero) | StructNeRF · NYUv2 | PSNR: 28.10 -> 24.03 | Hurt |
| add per-frame latent code (test code zero) | StructNeRF · ScanNet | PSNR: 24.67 -> 23.73 | Hurt |
| use StructNeRF instead of Dense Depth Priors (pretrained depth completion) | Dense Depth Priors · NYUv2 | PSNR: 26.34 -> 28.10 | Improved |
| use StructNeRF instead of Dense Depth Priors on ScanNet | Dense Depth Priors · ScanNet | Depth RMSE: 0.1929 -> 0.2298 | Hurt |
Closest papers by meaning
- Dense Depth Priors for Neural Radiance Fields From Sparse Input Views (CVPR 2022 · similarity 0.90)
- NeRFPrior: Learning Neural Radiance Field as a Prior for Indoor Scene Reconstruction (CVPR 2025 · similarity 0.90)
- SparseGNV: Generating Novel Views of Indoor Scenes with Sparse RGB-D Images (AAAI 2024 · similarity 0.89)
- Few-Shot Neural Radiance Fields under Unconstrained Illumination (AAAI 2024 · similarity 0.89)
- AltNeRF: Learning Robust Neural Radiance Field via Alternating Depth-Pose Optimization (AAAI 2024 · similarity 0.88)
- ColNeRF: Collaboration for Generalizable Sparse Input Neural Radiance Field (AAAI 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.
