Research map / Video / Video segmentation and tracking
Semi-Supervised Video Desnowing Network via Temporal Decoupling Experts and Distribution-Driven Contrastive Regularization
Hongtao Wu, Angelica I Aviles-Rivero, Yijun Yang, Jingjing Ren, Sixiang Chen, Haoyu Chen, Lei Zhu
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 Temporal Decoupling Experts module (replacing FFN) to the baseline (ablation) | SemiVDN (M1 -> M2) · RVSD (synthetic) test | PSNR: 24.41 -> 25.16 | Improved |
| add Temporal Decoupling Experts module (ablation) | SemiVDN (M1 -> M2) · RVSD (synthetic) test | LPIPS: 0.0932 -> 0.0822 | Improved |
| add semi-supervised training with unlabeled real data (mean teacher) (ablation) | SemiVDN (M2 -> M3) · RVSD (synthetic) test | PSNR: 25.16 -> 25.29 | Improved |
| add semi-supervised training with unlabeled real data (ablation) | SemiVDN (M2 -> M3) · Realsnow85 test | MUSIQ: 50.69 -> 51.05 | Improved |
| add Distribution-driven Contrastive Regularization (ablation) | SemiVDN (M3 -> full) · RVSD (synthetic) test | PSNR: 25.29 -> 25.68 | Improved |
| add Distribution-driven Contrastive Regularization (ablation) | SemiVDN (M3 -> full) · Realsnow85 test | MUSIQ: 51.05 -> 51.57 | Improved |
| use SemiVDN instead of SVDNet | video desnowing · RVSD (synthetic) test | PSNR: 25.06 -> 25.68 | Improved |
| use SemiVDN instead of SVDNet on real snowy videos | video desnowing · Realsnow85 test | MUSIQ: 50.78 -> 51.57 | Improved |
Closest papers by meaning
- Snow Removal in Video: A New Dataset and A Novel Method (ICCV 2023 · similarity 0.89)
- SnowMaster: Comprehensive Real-world Image Desnowing via MLLM with Multi-Model Feedback Optimization (CVPR 2025 · similarity 0.88)
- Video Desnowing and Deraining Based on Matrix Decomposition (CVPR 2017 · similarity 0.87)
- Genuine Knowledge from Practice: Diffusion Test-Time Adaptation for Video Adverse Weather Removal (CVPR 2024 · similarity 0.86)
- Semi-Supervised State-Space Model with Dynamic Stacking Filter for Real-World Video Deraining (CVPR 2025 · similarity 0.85)
- Learning Real-World Image De-weathering with Imperfect Supervision (AAAI 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.
