Research map / 3D vision / Face and shape
DebSDF: Delving Into the Details and Bias of Neural Indoor Scene Reconstruction
Yuting Xiao, Jingwei Xu, Zehao Yu, Shenghua Gao
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 uncertainty-guided prior filtering (ablation) | MonoSDF · ScanNet | F-score: 73.30 -> 76.82 | Improved |
| add uncertainty-guided prior filtering (ablation) | MonoSDF · ICL-NUIM | F-score: 67.39 -> 73.97 | Improved |
| add uncertainty-guided ray sampling (ablation) | DebSDF with prior filtering · ScanNet | F-score: 76.82 -> 77.12 | Improved |
| add uncertainty-guided ray sampling (ablation) | DebSDF with prior filtering · ICL-NUIM | F-score: 73.97 -> 76.14 | Improved |
| add uncertainty-guided smooth regularization (ablation) | DebSDF with filtering and sampling · ScanNet | F-score: 77.12 -> 77.30 | Improved |
| add uncertainty-guided smooth regularization (ablation) | DebSDF with filtering and sampling · ICL-NUIM | F-score: 76.14 -> 76.75 | Improved |
| add bias-aware SDF-to-density transformation (ablation) | DebSDF with three uncertainty modules · ScanNet | F-score: 77.30 -> 78.54 | Improved |
| add bias-aware SDF-to-density transformation (ablation) | DebSDF with three uncertainty modules · ICL-NUIM | F-score: 76.75 -> 77.59 | Improved |
| predict uncertainty with view-dependent color network instead of geometry network (ablation) | DebSDF · ScanNet | F-score: 73.2 -> 78.5 | Improved |
| fuse depth and normal uncertainty with lambda 0.9 instead of lambda 0 (depth only) (ablation) | DebSDF · ICL-NUIM | F-score: 75.67 -> 77.39 | Improved |
| use lambda 0.9 instead of 1 (normal only) (ablation) | DebSDF · ICL-NUIM | F-score: 77.16 -> 77.39 | Improved |
| replace SDF-to-density transformation with bias-aware one (ablation) | TUVR / NeuS · DTU (no prior) | Chamfer: 0.79 -> 0.75 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- HelixSurf — HelixSurf: A Robust and Efficient Neural Implicit Surface Learning of Indoor Scenes With Iterative Intertwined Regularization (CVPR 2023) · better on 2 of 2 shared results
- Unisurf — UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction (ICCV 2021) · better on 2 of 2 shared results
- NeuS — NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction (NeurIPS 2021) · 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
- ND-SDF: Learning Normal Deflection Fields for High-Fidelity Indoor Reconstruction (ICLR 2025 · similarity 0.92)
- NC-SDF: Enhancing Indoor Scene Reconstruction Using Neural SDFs with View-Dependent Normal Compensation (CVPR 2024 · similarity 0.91)
- HelixSurf: A Robust and Efficient Neural Implicit Surface Learning of Indoor Scenes With Iterative Intertwined Regularization (CVPR 2023 · similarity 0.90)
- Sparis: Neural Implicit Surface Reconstruction of Indoor Scenes from Sparse Views (AAAI 2025 · similarity 0.89)
- NeRFPrior: Learning Neural Radiance Field as a Prior for Indoor Scene Reconstruction (CVPR 2025 · similarity 0.89)
- MP-SfM: Monocular Surface Priors for Robust Structure-from-Motion (CVPR 2025 · 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.
