Research map / 3D vision / Gaussian splatting
Sparse-View Surface Reconstruction using Gaussian Splatting through High-Confidence Depth Propagation with Normal Priors
Liang Han, Bangcai Wei, Junsheng Zhou, Yushen Liu, Zhizhong Han
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 normal prior loss L_normal (ablation) | 3DGS rendering baseline · DTU ablation (3 views) | mean CD: 2.79 -> 2.28 | Improved |
| add opacity decay strategy (ablation) | 3DGS with L_normal · DTU ablation (3 views) | mean CD: 2.28 -> 1.55 | Improved |
| add depth propagation constraint L_dp (ablation) | 3DGS with L_normal and O.D. · DTU ablation (3 views) | mean CD: 1.55 -> 1.13 | Improved |
| add abnormal depth edge smoothing L_ds (ablation) | 3DGS with L_normal and O.D. · DTU ablation (3 views) | mean CD: 1.55 -> 1.38 | Improved |
| add L_ds on top of L_dp (ablation) | full method · DTU ablation (3 views) | mean CD: 1.13 -> 1.02 | Improved |
| add L_dp on top of L_ds (ablation) | full method · DTU ablation (3 views) | mean CD: 1.38 -> 1.02 | Improved |
| raise propagation iterations from 4 to 10 (ablation) | depth propagation · DTU propagation iterations | CD: 1.126 -> 1.021 | Improved |
| raise propagation iterations beyond 10 (ablation) | depth propagation · DTU propagation iterations | CD: 1.021 -> 1.032 | No clear effect |
| use Metric3Dv2 instead of Omnidata normal prior (ablation) | Ours · DTU normal prior | CD: 1.14 -> 1.02 | Improved |
| use Metric3Dv2 instead of StableNormal normal prior (ablation) | Ours · DTU normal prior | CD: 1.05 -> 1.02 | Improved |
| use MASt3R dense point cloud initialization | Ours · DTU (3 small-overlap views, 15 scenes) | mean CD: 1.02 -> 0.85 | Improved |
| use Ours instead of MAtCha | 3DGS sparse reconstruction · Tanks and Temples (4 scenes), 20 views | F1: 0.218 -> 0.293 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 7 of 7 on most shared results.
- FatesGS — FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency (AAAI 2025) · better on 4 of 4 shared results
- Sparse2DGS — Sparse2DGS: Geometry-Prioritized Gaussian Splatting for Surface Reconstruction from Sparse Views (CVPR 2025) · better on 1 of 1 shared result
- NeuSurf — NeuSurf: On-Surface Priors for Neural Surface Reconstruction from Sparse Input Views (AAAI 2024) · better on 1 of 1 shared result
- UFORecon — UFORecon: Generalizable Sparse-View Surface Reconstruction from Arbitrary and Unfavorable Sets (CVPR 2024) · better on 1 of 1 shared result
- SparseCraft — SparseCraft: Few-Shot Neural Reconstruction through Stereopsis Guided Geometric Linearization (ECCV 2024) · better on 1 of 1 shared result
- VolRecon — VolRecon: Volume Rendering of Signed Ray Distance Functions for Generalizable Multi-View Reconstruction (CVPR 2023) · better on 1 of 1 shared result
- ReTR — ReTR: Modeling Rendering Via Transformer for Generalizable Neural Surface Reconstruction (NeurIPS 2023) · 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
- FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency (AAAI 2025 · similarity 0.95)
- Sparse2DGS: Geometry-Prioritized Gaussian Splatting for Surface Reconstruction from Sparse Views (CVPR 2025 · similarity 0.95)
- TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformers (AAAI 2025 · similarity 0.91)
- Evolving High-Quality Rendering and Reconstruction in a Unified Framework with Contribution-Adaptive Regularization (CVPR 2025 · similarity 0.91)
- SurfelSplat: Learning Efficient and Generalizable Gaussian Surfel Representations for Sparse-View Surface Reconstruction (NeurIPS 2025 · similarity 0.90)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.90)
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.
