Research map / 3D vision / NeRF and radiance fields
SkyLume: A Large-Scale Multi-Illumination Aerial Benchmark for Urban Scene Reconstruction and Beyond
Zhuoxiao Li, Wenzong Ma, Taoyu Wu, Jinjing Zhu, Shuai Zhang, Jing OU, Tongyan Hua, Yinrui Ren, Rongjun Qin, Hui Xiong, Wufan Zhao
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 Ref-GS instead of GS-IR | inverse rendering · SkyLume cross-time albedo | TCC Overall mean: 0.721 -> 0.775 | Improved |
| use Ref-Gaussian instead of GS-IR | inverse rendering · SkyLume cross-time albedo | TCC Overall mean: 0.721 -> 0.658 | Hurt |
| use Ref-Gaussian instead of GS-IR | inverse rendering · SkyLume cross-time albedo | TCC-MAE mean: 0.700 -> 0.513 | Hurt |
| use CityGaussianV2 instead of GOF (opacity-based depth) | geometry · SkyLume cross-time geometry | TCC-Geometry F-1 at 0.25 m: 0.431 -> 0.547 | Improved |
| use alpha-blending TSDF depth (2DGS) instead of GOF | geometry · SkyLume cross-time geometry | TCC-Geometry F-1 at 0.25 m: 0.431 -> 0.570 | Improved |
| use geometry-oriented 2DGS instead of Abs-GS | novel view synthesis · SkyLume NVS Period 1 (Gym) | PSNR: 21.183 -> 20.350 | Hurt |
| use geometry-oriented PGSR instead of 3DGS | novel view synthesis · SkyLume NVS Period 1 (Gym) | LPIPS: 0.318 -> 0.424 | Hurt |
| use Abs-GS instead of 3DGS | novel view synthesis · SkyLume NVS Period 1 (Gym) | LPIPS: 0.318 -> 0.305 | Improved |
| use Octree-GS instead of 3DGS | novel view synthesis · SkyLume NVS Period 1 (Gym) | PSNR: 21.144 -> 20.931 | No clear effect |
| capture under direct sunlight instead of diffuse overcast | geometry methods · SkyLume geometry | qualitative | Hurt |
Closest papers by meaning
- UAVLight: A Benchmark for Illumination-Robust 3D Reconstruction in Unmanned Aerial Vehicle (UAV) Scenes (CVPR 2026 · similarity 0.92)
- AirZoo: A Unified Large-Scale Dataset for Grounding Aerial Geometric 3D Vision (ECCV 2026 · similarity 0.89)
- Olbedo: An Albedo and Shading Aerial Dataset for Large-Scale Outdoor Environments (CVPR 2026 · similarity 0.88)
- AerialMegaDepth: Learning Aerial-Ground Reconstruction and View Synthesis (CVPR 2025 · similarity 0.87)
- Seeing Across Skies and Streets: Feedforward 3D Reconstruction from Satellite, Drone, and Ground Images (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.
