Research map / 3D vision / Gaussian splatting
UniTriSplat: A Unified 3D Gaussian Splatting Framework with Uniform Spherical Rasterization for Universal Cameras
Yipeng Zhu, Huajian Huang, Tristan Braud, Sai Kit Yeung
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 HEALPix-SSIM loss (ablation) | UniTriSplat · OmniBlender | PSNR (dB): 26.42 -> 32.57 | Improved |
| add radian-space density control (ablation) | UniTriSplat · OmniBlender | PSNR (dB): 24.83 -> 32.57 | Improved |
| add multi-view consistent final pruning (ablation) | UniTriSplat · OmniBlender | PSNR (dB): 32.57 -> 28.27 | Hurt |
| use NESTED quadtree search (x4) instead of RING scan (x4) (ablation) | UniTriSplat · OmniBlender | PSNR (dB): 32.57 -> 33.64; time 28.2 -> 49.1 | Mixed |
| reduce quadtree search depth from x4 to x2 (ablation) | UniTriSplat · OmniBlender | PSNR (dB): 33.64 -> 26.49 | Hurt |
| reduce quadtree search depth from x4 to x1 (ablation) | UniTriSplat · OmniBlender | PSNR (dB): 33.64 -> 20.36 | Hurt |
| use higher HEALPix level Ours(H) instead of Ours(L) | UniTriSplat · 360Roam 6080x3040 | PSNR (dB): 21.82 -> 22.53 | Improved |
| use lower HEALPix level Ours(L) at low resolution | UniTriSplat · 360Roam 1520x760 | PSNR (dB): 26.10 -> 23.84 | Hurt |
| use UniTriSplat rasterizer instead of OP43DGS on fisheye | UniTriSplat · FIORD (fisheye) | PSNR (dB): 21.64 -> 24.27 | Improved |
| use UniTriSplat instead of OmniGS | UniTriSplat · OmniBlender | PSNR (dB): 33.21 -> 32.57 | Hurt |
| use UniTriSplat instead of vanilla 3DGS | UniTriSplat · Mip-NeRF 360 (perspective) | PSNR (dB): 25.59 -> 27.57 | Improved |
| render perspective views from UniTriSplat-trained scene rather than 3DGS-rendered | UniTriSplat · OmniBlender cross-camera | PSNR (dB): 22.54 -> 28.17 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ODGS — ODGS: 3D Scene Reconstruction from Omnidirectional Images with 3D Gaussian Splattings (NeurIPS 2024) · better on 3 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
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024 · similarity 0.91)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.90)
- UVGS: Reimagining Unstructured 3D Gaussian Splatting using UV Mapping (CVPR 2025 · similarity 0.90)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- Splatter Image: Ultra-Fast Single-View 3D Reconstruction (CVPR 2024 · similarity 0.90)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (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.
