AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Unpaired Learning for Deep Image Deraining With Rain Direction Regularizer

Yang Liu, Ziyu Yue, Jinshan Pan, Zhixun Su

ICCV 2021 · Label-efficient and robust learning › Semi-supervised learning · approach: Pseudo-labeling for specialized tasks · 61 citations (OpenAlex)

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 changedWhereMeasuredEffect
add the rain direction regularizer to rain map estimation (ablation)Ours · RainDirectionPSNR: 30.34 -> 31.35Improved
add the rain direction regularizer (lambda > 0 instead of 0) (ablation)Ours · validation set average PSNRAvg. PSNR: nearly 1.6 dBImproved
use 5x5 or 7x7 directional kernels instead of 3x3 (horizontal/vertical gradients) (ablation)rain direction regularizer · validation set average PSNRAvg. PSNRImproved
vary trade-off weight lambda over a wide range (ablation)rain direction regularizer · validation set average PSNRAvg. PSNRNo clear effect
add the knowledge distillation loss (ablation)Ours · Rain100LPSNR: 39.15 -> 40.54Improved
add the knowledge distillation loss (ablation)Ours · Rain100LLPIPS: 0.013 -> 0.009Improved
add the knowledge distillation loss (ablation)Ours · real rainy image examplevisual qualityImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy