Research map / 3D vision / Gaussian splatting
WeatherCity: Urban Scene Reconstruction with Controllable Multi-Weather Transformation
Wenhua Wu, Huai Guan, Zhe Liu, Hesheng Wang
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 weather Gaussian (shared features + weather decoders) (ablation) | WeatherCity · Ablation | CLIP-S: 0.781 -> 0.880 | Improved |
| add weather Gaussian (shared features + weather decoders) (ablation) | WeatherCity · Ablation | CLIP-DS: 0.212 -> 0.320 | Improved |
| add weather Gaussian (shared features + weather decoders) (ablation) | WeatherCity · Ablation | Sem-CS: 0.894 -> 0.943 | Improved |
| add content consistency loss (ablation) | WeatherCity · Ablation | CLIP-S: 0.817 -> 0.880 | Improved |
| add content consistency loss (ablation) | WeatherCity · Ablation | CLIP-DS: 0.289 -> 0.320 | Improved |
| add content consistency loss (ablation) | WeatherCity · Ablation | Sem-CS: 0.916 -> 0.943 | Improved |
| add all proposed modules on top of Qwen-Image editing (ablation) | Qwen-Image baseline · Ablation | CLIP-S: 0.735 -> 0.880 | Improved |
| lift 2D editing to 4D weather Gaussians | Qwen-Image · Waymo Open Dataset | Sem-CS: 0.843 -> 0.915 | Improved |
| lift 2D editing to 4D weather Gaussians | Qwen-Image · nuScenes | Sem-CS: 0.902 -> 0.968 | Improved |
| use WeatherCity instead of NeRF-based ClimateNeRF | ClimateNeRF · Waymo scene 788 | FPS: 0.032 -> 25.67 | Improved |
| use WeatherCity instead of ClimateNeRF | ClimateNeRF · Waymo scene 788 (snowy) | CLIP-S: 0.807 -> 0.847 | Improved |
| use WeatherCity instead of Qwen-Image for fog | Qwen-Image · Waymo Open Dataset (foggy) | CLIP-DS: 0.251 -> 0.278 | Improved |
Closest papers by meaning
- WeatherEdit: Controllable Weather Editing with 4D Gaussian Field (AAAI 2026 · similarity 0.92)
- IntrinsicWeather: Controllable Weather Editing in Intrinsic Space (CVPR 2026 · similarity 0.89)
- STORM: Spatio-TempOral Reconstruction Model For Large-Scale Outdoor Scenes (ICLR 2025 · similarity 0.87)
- Dynamic 3D Gaussian Fields for Urban Areas (NeurIPS 2024 · similarity 0.85)
- Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start (NeurIPS 2025 · similarity 0.85)
- Harnessing Joint Rain-/Detail-aware Representations to Eliminate Intricate Rains (ICLR 2024 · similarity 0.84)
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.
