Research map / 3D vision / Gaussian splatting
MuSASplat: Efficient Sparse-View 3D Gaussian Splats via Lightweight Multi-Scale Adaptation
Muyu Xu, Fangneng Zhan, Xiaoqin Zhang, Ling Shao, Shijian Lu
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 Multi-Scale Adapter to the frozen ViT (ablation) | MuSASplat · RE10K 2-view | PSNR: 20.12 -> 24.32 | Improved |
| use Multi-Scale Adapter instead of LoRA3D (ablation) | MuSASplat · RE10K 2-view | PSNR: 22.89 -> 24.32 | Improved |
| add Feature Fusion Aggregator (ablation) | MuSASplat · RE10K 2-view | PSNR: 23.50 -> 24.32 | Improved |
| use Feature Fusion Aggregator instead of memory bank (ablation) | MuSASplat · RE10K 2-view | training speed (it/s): 0.43 -> 1.81 | Improved |
| use Feature Fusion Aggregator instead of memory bank (ablation) | MuSASplat · RE10K 2-view | PSNR: 23.66 -> 24.32 | Improved |
| swap LoRA3D for MuSA adapter in NoPoSplat | NoPoSplat · RE10K | PSNR: 21.72 -> 23.39 | Improved |
| swap LoRA3D for MuSA adapter in SelfSplat | SelfSplat · RE10K | PSNR: 21.60 -> 23.14 | Improved |
| add mini-grid intra-patch branch to MuSA (ablation) | MuSA adapter · all tested datasets | rendering quality | No clear effect |
| use MuSASplat (frozen backbone, 119M) instead of full fine-tuning NoPoSplat (612M) | pose-free 3DGS · RE10K 2-view | PSNR: 24.833 -> 24.324 | Hurt |
| use MuSASplat feature aggregator in 5-view setting versus memory-bank baselines | pose-free 3DGS · RE10K 5-view | PSNR: 22.181 -> 24.721 (vs PREF3R) | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- SelfSplat — SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025) · better on 1 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
- Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis (CVPR 2024 · similarity 0.91)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.91)
- Generative Sparse-View Gaussian Splatting (CVPR 2025 · similarity 0.91)
- MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Models (CVPR 2025 · similarity 0.91)
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024 · similarity 0.91)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (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.
