Research map / 3D vision / Gaussian splatting
UAV4D: Dynamic Neural Rendering of Human-Centric UAV Imagery Using Gaussian Splatting
Jaehoon Choi, Dongki Jung, Chris Maxey, Sungmin Eum, Yonghan Lee, Dinesh Manocha, Heesung Kwon
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 SMPL human Gaussian splats (ablation) | UAV4D · VisDrone | PSNR: 24.33 -> 26.03 | Improved |
| add SMPL human Gaussian splats (ablation) | UAV4D · VisDrone | LPIPS: 0.212 -> 0.147 | Improved |
| add scale optimization for human mesh placement (ablation) | UAV4D · VisDrone | PSNR: 23.02 -> 26.03 | Improved |
| add scale optimization for human mesh placement (ablation) | UAV4D · VisDrone | SSIM: 0.651 -> 0.800 | Improved |
| add human mesh refinement (ablation) | UAV4D · VisDrone | PSNR: 25.79 -> 26.03 | Improved |
| add human mesh refinement (ablation) | UAV4D · VisDrone | LPIPS: 0.165 -> 0.147 | Improved |
| use UAV4D instead of 3DGS | full image · VisDrone | PSNR: 24.59 -> 26.03 | Improved |
| use UAV4D instead of Deformable-GS | full image · Manipal-UAV | PSNR: 31.22 -> 30.94 | Hurt |
| use UAV4D instead of TK-Planes | human-only regions · Okutama-Action | PSNR: 19.14 -> 19.49 | Improved |
| use UAV4D instead of 4DGS | human-only regions · Manipal-UAV | PSNR: 19.23 -> 19.38 | Improved |
| use UAV4D instead of 3DGS | full image · Okutama-Action | LPIPS: 0.228 -> 0.184 | Improved |
| use UAV4D instead of Deformable-GS | full image · VisDrone | PSNR: 15.46 -> 26.03 | Improved |
Closest papers by meaning
- DroneSplat: 3D Gaussian Splatting for Robust 3D Reconstruction from In-the-Wild Drone Imagery (CVPR 2025 · similarity 0.88)
- Horizon-GS: Unified 3D Gaussian Splatting for Large-Scale Aerial-to-Ground Scenes (CVPR 2025 · similarity 0.87)
- Guess The Unseen: Dynamic 3D Scene Reconstruction from Partial 2D Glimpses (CVPR 2024 · similarity 0.87)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.87)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.86)
- PKU-DyMVHumans: A Multi-View Video Benchmark for High-Fidelity Dynamic Human Modeling (CVPR 2024 · 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.
