Research map / Label-efficient and robust learning / Domain adaptation and generalization
Adverse Weather Optical Flow: Cumulative Homogeneous-Heterogeneous Adaptation
Hanyu Zhou, Yi Chang, Zhiwei Shi, Wending Yan, Gang Chen, Yonghong Tian, Luxin Yan
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 knowledge transfer via implicit encoder features (ablation) | CH2DA-Flow framework · Weather-GOF | EPE: 2.94 -> 1.21 | Improved |
| add homogeneous adaptation (ablation) | CH2DA-Flow framework · Weather-GOF | EPE: 1.21 -> 0.82 | Improved |
| add heterogeneous adaptation (ablation) | CH2DA-Flow framework · Weather-GOF | EPE: 1.21 -> 1.03 | Improved |
| combine homogeneous and heterogeneous adaptation (ablation) | CH2DA-Flow framework · Weather-GOF | EPE: 0.82 -> 0.72 | Improved |
| combine homogeneous and heterogeneous adaptation (ablation) | CH2DA-Flow framework · DenseFog | EPE: 2.94 -> 2.82 | Improved |
| add geometric rigid-flow loss alone (ablation) | CH2DA-Flow · Weather-GOF | EPE: 2.94 -> 2.87 | No clear effect |
| add warp error contrastive loss (ablation) | CH2DA-Flow · Weather-GOF | EPE: 1.35 -> 1.20 | Improved |
| add KL correlation alignment loss (ablation) | CH2DA-Flow · Weather-GOF | EPE: 1.09 -> 0.72 | Improved |
| add pseudo-label self-supervised loss (ablation) | CH2DA-Flow · Weather-GOF | EPE: 0.88 -> 0.72 | Improved |
| use edge-aware and entropy-aware sampling instead of random sampling (ablation) | contrastive learning · Rain-KITTI2015 | EPE: 5.83 -> 5.35 | Improved |
| use edge-aware sampling for clean patches only (random negatives) (ablation) | contrastive learning · Rain-GOF | EPE: 0.87 -> 0.76 | Improved |
| use FlowFormer instead of RAFT backbone (ablation) | CH2DA-Flow · Weather-GOF | EPE: 0.75 -> 0.72 | Improved |
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.
- Selflow — SelFlow: Self-Supervised Learning of Optical Flow (CVPR 2019) · better on 1 of 1 shared result
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
- Unsupervised Hierarchical Domain Adaptation for Adverse Weather Optical Flow (AAAI 2023 · similarity 0.95)
- Exploring the Common Appearance-Boundary Adaptation for Nighttime Optical Flow (ICLR 2024 · similarity 0.86)
- Parsing All Adverse Scenes: Severity-Aware Semantic Segmentation with Mask-Enhanced Cross-Domain Consistency (AAAI 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)
- End-to-End Video Semantic Segmentation in Adverse Weather using Fusion Blocks and Temporal-Spatial Teacher-Student Learning (NeurIPS 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.
