Research map / 3D vision / Gaussian splatting
Hierarchy UGP: Hierarchy Unified Gaussian Primitive for Large-Scale Dynamic Scene Reconstruction
Hongyang Sun, Qinglin Yang, Jiawei Wang, Zhen Xu, Chen Liu, Yida Wang, Kun Zhan, Hujun Bao, Xiaowei Zhou, Sida Peng
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 Block-wise Objects training (ablation) | Hierarchy UGP · Waymo vehicle regions | PSNR+: 21.51 -> 23.15 | Improved |
| add Block-wise Objects training (ablation) | Hierarchy UGP · Waymo overall | PSNR: 28.58 -> 28.64 | No clear effect |
| add temporal scale initialization (ablation) | Hierarchy UGP · Waymo human regions | PSNR: 23.22 -> 34.93 | Improved |
| add temporal scale initialization (ablation) | Hierarchy UGP · Waymo human regions | SSIM: 0.791 -> 0.943 | Improved |
| add virtual warping view supervision (ablation) | Hierarchy UGP · Dynamic City lane shift 2m NVS | visual quality | Improved |
| use Hierarchy UGP instead of Hierarchical GS | Gaussian scene model · Dynamic City Large 001 | PSNR: 26.74 -> 27.01 | Improved |
| use Hierarchy UGP instead of OmniRe | dynamic scene model · Dynamic City Large 002 | PSNR: 22.17 -> 25.25 | Improved |
| use Hierarchy UGP instead of 4D-GS | dynamic scene model · Dynamic City Sub-scene 002 | PSNR: 25.51 -> 27.61 | Improved |
| use Hierarchy UGP instead of Street GS for extrapolation | dynamic scene model · Dynamic City Large 001 lane shift | FID: 90.99 -> 68.27 | Improved |
| use Hierarchy UGP instead of Street GS | dynamic scene model · Waymo scene 023 reconstruction pedestrian | PSNR*: 24.75 -> 34.10 | Improved |
| use Hierarchy UGP instead of Street GS for novel view synthesis | dynamic scene model · Waymo scene 023 NVS pedestrian | PSNR*: 21.79 -> 21.64 | Hurt |
| use Hierarchy UGP instead of OmniRe for novel view synthesis | dynamic scene model · Waymo scene 023 NVS pedestrian | PSNR*: 23.07 -> 21.64 | Hurt |
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.
- OmniRe — OmniRe: Omni Urban Scene Reconstruction (ICLR 2025) · better on 6 of 6 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
- Dynamic 3D Gaussian Fields for Urban Areas (NeurIPS 2024 · similarity 0.88)
- FreeTimeGS: Free Gaussian Primitives at Anytime Anywhere for Dynamic Scene Reconstruction (CVPR 2025 · similarity 0.87)
- Multi-Level Neural Scene Graphs for Dynamic Urban Environments (CVPR 2024 · similarity 0.87)
- OmniRe: Omni Urban Scene Reconstruction (ICLR 2025 · similarity 0.87)
- World from Motion: Generative Dynamic Gaussian Reconstruction from Monocular Video (NeurIPS 2026 · similarity 0.86)
- Horizon-GS: Unified 3D Gaussian Splatting for Large-Scale Aerial-to-Ground Scenes (CVPR 2025 · similarity 0.86)
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.
