Research map / 3D vision / Gaussian splatting
DPGS: A Diffusion-Prior Guided Framework for Large-Scale 3D Gaussian Splatting Reconstruction
Shidong Zhang, shuaixin li, Juntong Qi, Haoxin Zhang, Xiao zhang, Xiaozhou Zhu, Wen Yao
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 dense initialization (DI) (ablation) | DPGS · Self-collected | PSNR: 21.15 -> 21.43 | Improved |
| add dense initialization (DI) (ablation) | DPGS · Self-collected | SSIM: 0.458 -> 0.474 | Improved |
| add diffusion supervision (DS) (ablation) | DPGS · Self-collected | FID: 72.37 -> 48.00 | Improved |
| add diffusion supervision (DS) (ablation) | DPGS · Self-collected | LPIPS: 0.441 -> 0.428 | Improved |
| add DS on top of DI (ablation) | DPGS · Self-collected | PSNR: 21.43 -> 21.34 | Hurt |
| add DS on top of DI (ablation) | DPGS · Self-collected | FID: 63.34 -> 43.53 | Improved |
| add DS on top of DI (ablation) | DPGS · Self-collected | LPIPS: 0.436 -> 0.415 | Improved |
| use RoMa instead of ASpanFormer for dense matching (ablation) | DPGS · Self-collected | PSNR: 21.38 -> 21.43 | Improved |
| use RoMa instead of UFM for dense matching (ablation) | DPGS · Self-collected | PSNR: 21.30 -> 21.43 | Improved |
| use SD-Turbo instead of DiffusionSat (ablation) | DPGS diffusion prior · Self-collected | FID: 82.36 -> 43.53 | Improved |
| use SD-Turbo instead of FlowEdit (ablation) | DPGS diffusion prior · Self-collected | FID: 58.20 -> 43.53 | Improved |
| DPGS versus Momentum-GS | Large-scale 3DGS · MatrixCity | LPIPS: 0.180 -> 0.115 | 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.
- DOGS — DOGS: Distributed-Oriented Gaussian Splatting for Large-Scale 3D Reconstruction Via Gaussian Consensus (NeurIPS 2024) · better on 3 of 5 shared results
- Momentum-GS — Momentum-GS: Momentum Gaussian Self-Distillation for High-Quality Large Scene Reconstruction (ICCV 2025) · 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
- SparseSplat: Towards Applicable Feed-Forward 3D Gaussian Splatting with Pixel-Unaligned Prediction (CVPR 2026 · similarity 0.91)
- Evolving High-Quality Rendering and Reconstruction in a Unified Framework with Contribution-Adaptive Regularization (CVPR 2025 · similarity 0.90)
- Taming Video Diffusion Prior with Scene-Grounding Guidance for 3D Gaussian Splatting from Sparse Inputs (CVPR 2025 · similarity 0.90)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.89)
- CompactSplat: Spatially Adaptive Gaussian Distribution for Feedforward Scene Reconstruction (NeurIPS 2026 · similarity 0.89)
- Wonderland: Navigating 3D Scenes from a Single Image (CVPR 2025 · similarity 0.89)
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.
