Research map / 3D vision / Face and shape
NeuRIS: Neural Reconstruction of Indoor Scenes Using Normal Priors
Jiepeng Wang, Peng Wang, Xiaoxiao Long, Christian Theobalt, Taku Komura, Lingjie Liu, Wenping 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 |
|---|---|---|---|
| add monocular normal prior supervision (ablation) | NeuS · ScanNet ablation (8 scenes) | F-score: 0.284 -> 0.724 | Improved |
| add geometric (photometric-consistency) check of normal priors (ablation) | NeuS with normal priors · ScanNet ablation (8 scenes) | F-score: 0.724 -> 0.736 | Improved |
| add normal priors (ablation) | NeuS · ScanNet ablation (8 scenes) | Accuracy error: 0.183 -> 0.050 | Improved |
| add geometric check of normal priors (ablation) | NeuS with normal priors · ScanNet ablation (8 scenes) | Accuracy error: 0.050 -> 0.046 | Improved |
| add geometric check of normal priors (ablation) | NeuS with normal priors · ScanNet ablation (8 scenes) | Completeness error: 0.053 -> 0.053 | No clear effect |
| NeuRIS versus NeuS on ScanNet | NeuS · ScanNet room-scale reconstruction (8 scenes) | F-score: 0.284 -> 0.736 | Improved |
| NeuRIS versus NeuralRecon on ScanNet | NeuralRecon · ScanNet room-scale reconstruction (8 scenes) | F-score: 0.640 -> 0.736 | Improved |
| NeuRIS versus DeepV2D on completeness | DeepV2D · ScanNet room-scale reconstruction (8 scenes) | Completeness error: 0.049 -> 0.053 | Hurt |
| render normals with NeuRIS instead of monocular TiltedSN prediction | TiltedSN · ScanNet normal estimation | Mean angular error: 15.4 -> 14.7 | Improved |
| NeuRIS versus NeuS on novel view synthesis | NeuS · ScanNet novel view synthesis (500 novel views, 8 scenes) | PSNR: 22.7 -> 24.4 | Improved |
| NeuRIS versus NeRF on novel view synthesis | NeRF · ScanNet novel view synthesis (500 novel views, 8 scenes) | PSNR: 23.3 -> 24.4 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- NeuralRecon — NeuralRecon: Real-Time Coherent 3D Reconstruction From Monocular Video (CVPR 2021) · better on 3 of 4 shared results
- NeuS — NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction (NeurIPS 2021) · better on 4 of 4 shared results
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
- NeRFPrior: Learning Neural Radiance Field as a Prior for Indoor Scene Reconstruction (CVPR 2025 · similarity 0.88)
- UniSDF: Unifying Neural Representations for High-Fidelity 3D Reconstruction of Complex Scenes with Reflections (NeurIPS 2024 · similarity 0.87)
- ND-SDF: Learning Normal Deflection Fields for High-Fidelity Indoor Reconstruction (ICLR 2025 · similarity 0.87)
- Sparis: Neural Implicit Surface Reconstruction of Indoor Scenes from Sparse Views (AAAI 2025 · similarity 0.87)
- SC-NeuS: Consistent Neural Surface Reconstruction from Sparse and Noisy Views (AAAI 2024 · similarity 0.85)
- Global and Hierarchical Geometry Consistency Priors for Few-shot NeRFs in Indoor Scenes (CVPR 2024 · similarity 0.85)
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.
