Research map / 3D vision / Gaussian splatting
SurfSplat: Conquering Feedforward 2D Gaussian Splatting with Surface Continuity Priors
Bing He, Jingnan Gao, Yunuo Chen, Ning Cao, Gang Chen, Zhengxue Cheng, Li Song, Wenjun 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 forced alpha blending (on top of surface continuity prior) (ablation) | SurfSplat Ours-S · RealEstate10K 1024x1024 HRRC | PSNR: 17.576 -> 24.535 | Improved |
| add forced alpha blending (ablation) | SurfSplat Ours-S · RealEstate10K 256x256 standard | PSNR: 26.481 -> 27.001 | Improved |
| add surface continuity prior (and FAB) (ablation) | SurfSplat Ours-S · RealEstate10K 1024x1024 HRRC | PSNR: 18.563 -> 24.535 | Improved |
| add surface continuity prior (and FAB) (ablation) | SurfSplat Ours-S · RealEstate10K 256x256 standard | PSNR: 26.925 -> 27.001 | No clear effect |
| use 2DGS with prior instead of 3DGS DepthSplat with same backbone | SurfSplat Ours-L · RealEstate10K 1024x1024 HRRC | PSNR: 16.385 -> 24.897 | Improved |
| use SurfSplat vs DepthSplat at standard resolution | SurfSplat Ours-L · RealEstate10K 256x256 | PSNR: 27.504 -> 27.537 | No clear effect |
| scale monocular backbone from ViT-S to ViT-L (ablation) | SurfSplat · RealEstate10K 1024x1024 HRRC | PSNR: 24.535 -> 24.897 | Improved |
| vary tau_alpha to 0.3 (ablation) | SurfSplat Ours-S · RealEstate10K | PSNR: 27.001 -> 26.921 | No clear effect |
| vary tau_opa to 0.4 (ablation) | SurfSplat Ours-S · RealEstate10K | PSNR: 27.001 -> 26.992 | No clear effect |
| compare SurfSplat to DepthSplat cross-dataset | SurfSplat Ours · DL3DV (trained on RE10K) | PSNR: 28.141 -> 27.384 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- DepthSplat — DepthSplat: Connecting Gaussian Splatting and Depth (CVPR 2025) · better on 4 of 6 shared results
- HiSplat — HiSplat: Hierarchical 3D Gaussian Splatting for Generalizable Sparse-View Reconstruction (ICLR 2025) · better on 2 of 3 shared results
- MVSplat — MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images (ECCV 2024) · better on 3 of 3 shared results
- pixelSplat — pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024) · 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
- CompactSplat: Spatially Adaptive Gaussian Distribution for Feedforward Scene Reconstruction (NeurIPS 2026 · similarity 0.93)
- SurfelSplat: Learning Efficient and Generalizable Gaussian Surfel Representations for Sparse-View Surface Reconstruction (NeurIPS 2025 · similarity 0.92)
- Evolving High-Quality Rendering and Reconstruction in a Unified Framework with Contribution-Adaptive Regularization (CVPR 2025 · similarity 0.91)
- Geometry Field Splatting with Gaussian Surfels (CVPR 2025 · similarity 0.91)
- Sparse2DGS: Geometry-Prioritized Gaussian Splatting for Surface Reconstruction from Sparse Views (CVPR 2025 · similarity 0.90)
- FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency (AAAI 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.
