Research map / 3D vision / Gaussian splatting
MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images
Yuedong Chen, Haofei Xu, Chuanxia Zheng, Bohan Zhuang, Marc Pollefeys, Andreas Geiger, Tat-Jen Cham, Jianfei Cai
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 cost volume (plane-sweep feature matching) for depth (ablation) | MVSplat base · RealEstate10K | PSNR: 22.83 -> 26.12 | Improved |
| add cross-view attention in the Transformer (ablation) | MVSplat base · RealEstate10K | PSNR: 25.19 -> 26.12 | Improved |
| add cost volume refinement U-Net (ablation) | MVSplat base · RealEstate10K | PSNR: 25.45 -> 26.12 | Improved |
| add depth refinement U-Net (ablation) | MVSplat base · RealEstate10K | PSNR: 26.12 -> 26.39 | Improved |
| replace pixelSplat depth probability branch with MVSplat cost volume depth (ablation) | pixelSplat · RealEstate10K | PSNR: 25.89 -> 26.63 | Improved |
| use Swin Transformer instead of Epipolar Transformer (ablation) | MVSplat · RealEstate10K | PSNR: 26.09 -> 26.12 PSNR; time 0.055 -> 0.038 s | No clear effect |
| predict 3 Gaussians per pixel instead of 1 (ablation) | MVSplat · RealEstate10K | PSNR: 26.39 -> 26.54 PSNR; render 0.0015 -> 0.0023 s | Mixed |
| random init instead of UniMatch pretrained init (300K) (ablation) | MVSplat · RealEstate10K | PSNR: 26.39 -> 26.01 | Hurt |
| train MVSplat instead of pixelSplat for zero-shot transfer to DTU | RealEstate10K-trained models · DTU (zero-shot) | LPIPS: 0.560 -> 0.385 | Improved |
| increase input views from 2 to 3 at test time | pixelSplat · DTU (zero-shot) | PSNR: 12.89 -> 12.52 | Hurt |
| increase input views from 2 to 3 at test time | MVSplat · DTU (zero-shot) | PSNR: 13.94 -> 14.30 | Improved |
Later papers that compared with this method
65 later accepted papers report MVSplat as a baseline in their results tables; 62 of them report a better number on most of the results they share with it.
- MeshSplat: Generalizable Sparse-View Surface Reconstruction via Gaussian Splatting (AAAI 2026 · better on 10 of 10 shared results)
- SRSplat: Feed-Forward Super-Resolution Gaussian Splatting from Sparse Multi-View Images (AAAI 2026 · better on 9 of 9 shared results)
- Targeted Structure Completion for Sparse-View 3D Reconstruction in Autonomous Driving (ECCV 2026 · better on 9 of 9 shared results)
- TokenGS: Decoupling 3D Gaussian Prediction from Pixels with Learnable Tokens (CVPR 2026 · better on 6 of 7 shared results)
- FreeSplat++: Generalizable 3D Gaussian Splatting for Efficient Indoor Scene Reconstruction (TPAMI 2026 · better on 7 of 7 shared results)
- Aligned Novel View Image and Geometry Synthesis via Cross-modal Attention Instillation (ICLR 2026 · better on 6 of 6 shared results)
- ReconSplat: Generalizable 3D Scene Reconstruction Beyond Observed Views (ECCV 2026 · better on 5 of 6 shared results)
- UniSplat: Unified Spatio-Temporal Fusion via 3D Latent Scaffolds for Dynamic Driving Scene Reconstruction (ICLR 2026 · better on 5 of 5 shared results)
57 more not listed.
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.
- pixelSplat — pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024) · better on 10 of 10 shared results
- MuRF — MuRF: Multi-Baseline Radiance Fields (CVPR 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
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024 · similarity 0.91)
- SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images (ICLR 2025 · similarity 0.91)
- CompactSplat: Spatially Adaptive Gaussian Distribution for Feedforward Scene Reconstruction (NeurIPS 2026 · 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)
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.
