Research map / 3D vision / Depth and stereo
PanoSplatt3R: Leveraging Perspective Pretraining for Generalized Unposed Wide-Baseline Panorama Reconstruction
Jiahui Ren, Mochu Xiang, Jiajun Zhu, Yuchao 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 |
|---|---|---|---|
| use unposed PanoSplatt3R instead of posed panorama methods | PanoSplatt3R · HM3D | PSNR: Splatter-360 28.308 -> 28.938 | Improved |
| use PanoSplatt3R instead of Splatter-360 on SSIM | PanoSplatt3R · HM3D | SSIM: 0.875 -> 0.869 | Hurt |
| feed posed methods with SIFT+8PA estimated poses | Splatter-360 · HM3D | PSNR: 28.308 -> 14.765 | Hurt |
| initialize from MASt3R perspective pretrained weights instead of random (ablation) | PanoSplatt3R · HM3D | PSNR: 23.331 -> 28.200 | Improved |
| add RoPE rolling (ablation) | PanoSplatt3R · HM3D | PSNR: 27.916 -> 28.200 | Improved |
| use RoPE rolling instead of setting max rotation angle to 2pi (ablation) | PanoSplatt3R · HM3D | PSNR: 26.849 -> 28.200 | Improved |
| add two-stage progressive training (centers first) (ablation) | PanoSplatt3R · HM3D | PSNR: 27.217 -> 28.200 | Improved |
| use 2DGS instead of 3DGS (ablation) | PanoSplatt3R · HM3D | PSNR: 28.049 -> 28.200 | Improved |
| apply uniform scale recovery to the reconstruction (ablation) | PanoSplatt3R · Replica | PSNR: 28.181 -> 31.522 | Improved |
| compare to fixed-baseline PanSplat at narrow baseline (1.0m) | PanoSplatt3R · Matterport3D (1.0m) | WS-PSNR: 30.01 -> 27.47 | Hurt |
| compare to fixed-baseline PanSplat at wider baseline (2.0m) | PanoSplatt3R · Matterport3D (2.0m) | WS-PSNR: 21.19 -> 23.65 | Improved |
Later papers that compared with this method
1 later accepted paper reports PanoSplatt3R as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- OmniLife360: A Benchmark for 3D Reconstruction from In-the-Wild 360° Captures (ECCV 2026 · better on 1 of 1 shared result)
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.
- PanSplat — PanSplat: 4K Panorama Synthesis with Feed-Forward 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
- Splatter-360: Generalizable 360 Gaussian Splatting for Wide-baseline Panoramic Images (CVPR 2025 · similarity 0.90)
- Unified Panoramic Geometry Estimation via Multi-View Foundation Models (NeurIPS 2026 · similarity 0.89)
- SPARS3R: Semantic Prior Alignment and Regularization for Sparse 3D Reconstruction (CVPR 2025 · similarity 0.88)
- Scene4U: Hierarchical Layered 3D Scene Reconstruction from Single Panoramic Image for Your Immerse Exploration (CVPR 2025 · similarity 0.88)
- PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis (ICML 2025 · similarity 0.88)
- DepthMaster: Unified Monocular Depth Estimation for Perspective and Panoramic Images (NeurIPS 2026 · similarity 0.88)
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.
