Research map / 3D vision / Gaussian splatting
QuickSplat: Fast 3D Surface Reconstruction via Learned Gaussian Initialization
Yueh-Cheng Liu, Lukas Höllein, Matthias Nießner, Angela Dai
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 the learned initializer network to the optimizer-only pipeline (ablation) | QuickSplat · ScanNet++ | Abs err: 0.1332 -> 0.0581 | Improved |
| predict full 2DGS attributes in initializer instead of occupancy only (ablation) | QuickSplat · ScanNet++ | Abs err: 0.0897 -> 0.0581 | Improved |
| add the learned densifier network (ablation) | QuickSplat · ScanNet++ | Abs err: 0.0581 -> 0.0578 | Improved |
| initializer predicts more Gaussian attributes (color only to color, opacity, scales, rotation) (ablation) | QuickSplat initializer · ScanNet++ | Abs err: 0.0627 -> 0.0578 | Improved |
| add 2000 SGD refinement steps after learned optimization | QuickSplat · ScanNet++ | Acc (2cm): 0.5263 -> 0.5783 | Improved |
| use learned optimizer and densifier instead of 2DGS optimization and densification with same initializer (ablation) | QuickSplat · ScanNet++ | Acc (2cm): Acc (2cm) 0.4796 -> 0.5263; Abs err 0.0668 -> 0.0732 | Mixed |
| use QuickSplat initializer in SAGS instead of SfM initialization | SAGS · ScanNet++ | Abs err: 0.1292 -> 0.0692 | Improved |
| increase learned optimizer steps from T=0 to T=5 (ablation) | QuickSplat · ScanNet++ | Abs err: 0.0921 -> 0.0732 | Improved |
| increase learned optimizer steps from T=0 to T=5 (ablation) | QuickSplat · ScanNet++ | Chamfer: 0.1478 -> 0.1461 | No clear effect |
| replace 2DGS with QuickSplat on unseen ARKitScenes without fine-tuning | QuickSplat · ARKitScenes (5 scenes) | Abs err: 0.6978 -> 0.1775 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GS2Mesh — GS2Mesh: Surface Reconstruction from Gaussian Splatting via Novel Stereo Views (ECCV 2024) · better on 2 of 2 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
- CompactSplat: Spatially Adaptive Gaussian Distribution for Feedforward Scene Reconstruction (NeurIPS 2026 · similarity 0.91)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.91)
- Evolving High-Quality Rendering and Reconstruction in a Unified Framework with Contribution-Adaptive Regularization (CVPR 2025 · similarity 0.90)
- MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Models (CVPR 2025 · similarity 0.90)
- IndoorGS: Geometric Cues Guided Gaussian Splatting for Indoor Scene Reconstruction (CVPR 2025 · similarity 0.90)
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024 · 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.
