Research map / 3D vision / Gaussian splatting
MVPGS: Excavating Multi-view Priors for Gaussian Splatting from Sparse Input Views
Wangze Xu, Huachen Gao, Shihe Shen, Rui Peng, Jianbo Jiao, Ronggang Wang
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 forward-warping appearance constraint (L_fwd) for unseen views (ablation) | MVPGS · LLFF 3-view (res 8x) | PSNR: 17.32 -> 20.39 | Improved |
| add consistent structure regularization L_CS (ablation) | MVPGS · LLFF 3-view (res 8x) | PSNR: 20.06 -> 20.39 | Improved |
| add monocular depth rank loss L_mono (ablation) | MVPGS · LLFF 3-view (res 8x) | PSNR: 20.17 -> 20.39 | Improved |
| initialize 3DGS with MVS point clouds instead of COLMAP points (ablation) | 3DGS · LLFF 3-view (res 8x) | PSNR: 14.87 -> 16.16 | Improved |
| replace COLMAP init with MVS init alone, without other components | 3DGS · LLFF 3-view | PSNR: 14.87 -> 16.16 | Mixed |
| swap MVS pretraining data (DTU, BlendedMVS, DTU+BlendedMVS ft) (ablation) | MVPGS · LLFF 3-view (res 8x) | PSNR: 20.39 / 20.17 / 20.34 | No clear effect |
| apply MVPGS versus FreeNeRF/SparseNeRF on T&T large-scale scenes | MVPGS · T&T 3-view | PSNR: 20.62 -> 25.57 | Improved |
| use MVPGS in 2-view extremely sparse setting | MVPGS · LLFF 2-view (res 8x) | PSNR: 17.44 -> 18.53 | Improved |
| increase to 5 input views on LLFF | MVPGS · LLFF 5-view (res 8x) | PSNR: 22.18 vs FreeNeRF 22.57 | Mixed |
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.
- SparseNeRF — SparseNeRF: Distilling Depth Ranking for Few-shot Novel View Synthesis (ICCV 2023) · better on 4 of 4 shared results
- FreeNeRF — FreeNeRF: Improving Few-Shot Neural Rendering With Free Frequency Regularization (CVPR 2023) · better on 2 of 3 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
- FewViewGS: Gaussian Splatting with Few View Matching and Multi-stage Training (NeurIPS 2024 · similarity 0.92)
- NexusGS: Sparse View Synthesis with Epipolar Depth Priors in 3D Gaussian Splatting (CVPR 2025 · similarity 0.92)
- 3DGS-Enhancer: Enhancing Unbounded 3D Gaussian Splatting with View-consistent 2D Diffusion Priors (NeurIPS 2024 · similarity 0.91)
- LP-3DGS: Learning to Prune 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.91)
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.91)
- Quantifying and Alleviating Co-Adaptation in Sparse-View 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.91)
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.
