Research map / 3D vision / Depth and stereo
Weather-Conditioned Depth Anything
Zhaoming Xu, Chan-Wei Hu, Kuan-Ru Huang, Zihao Zhu, Renjie Li, Yang Zhou, Zhengzhong Tu
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 |
|---|---|---|---|
| remove distillation loss L_dis (ablation) | DA-W · pooled weather benchmarks | AbsRel: 0.117 -> 0.352 | Hurt |
| add pair loss L_pair (ablation) | DA-W · pooled weather benchmarks | AbsRel: 0.120 -> 0.117 | Improved |
| add augmentation consistency loss L_aug (ablation) | DA-W · pooled weather benchmarks | AbsRel: 0.116 -> 0.117 | No clear effect |
| use weather-conditioned modulation instead of RGB restoration head (ablation) | DA-W · pooled weather benchmarks | AbsRel: 0.118 vs 0.117 | No clear effect |
| decoder-only tuning instead of full fine-tuning (ablation) | DA-W · NuScenes-night | delta1: 0.727 -> 0.741 | Improved |
| add real-world degradation data (ablation) | DA-W · RobotCar-night | AbsRel: 0.248 -> 0.243 | Improved |
| add AdaLN weather injection (ablation) | DA-W · RobotCar-night | AbsRel: 0.243 -> 0.239 | Improved |
| add AdaLN weather injection (ablation) | DA-W · DS-fog | AbsRel: 0.100 -> 0.101 | Hurt |
| full fine-tuning of encoder instead of decoder-only (ablation) | DA-W · pooled weather benchmarks | AbsRel: 0.119 vs 0.117 | No clear effect |
| use DA-W instead of DA v2 | DA-W · KITTI-C | AbsRel: Snow 0.115 -> 0.107 | Improved |
| use DA-W instead of DA v2 | DA-W · KITTI mix-weather | AbsRel: Fog+Rain+Snow+Night 0.268 -> 0.236 | Improved |
| use DA-W instead of DA v2/DA-AC | DA-W · RobotCar-night | AbsRel: 0.239 vs 0.239 (DA v3 0.169) | No clear effect |
Closest papers by meaning
- Self-supervised Monocular Depth Estimation: Let's Talk About The Weather (ICCV 2023 · similarity 0.89)
- Depth-Aware Blind Image Decomposition for Real-World Adverse Weather Recovery (ECCV 2024 · similarity 0.89)
- Synthetic-to-Real Self-supervised Robust Depth Estimation via Learning with Motion and Structure Priors (CVPR 2025 · similarity 0.88)
- Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data (CVPR 2024 · similarity 0.86)
- Real-World Adverse Weather Image Restoration via Dual-Level Reinforcement Learning with High-Quality Cold Start (NeurIPS 2025 · similarity 0.85)
- Genuine Knowledge from Practice: Diffusion Test-Time Adaptation for Video Adverse Weather Removal (CVPR 2024 · similarity 0.85)
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.
