AI Paper Atlas

Research map / Image generation / Image restoration

Not Just Streaks: Towards Ground Truth for Single Image Deraining

Yunhao Ba, Howard Zhang, Ethan Yang, Akira Suzuki, Arnold Pfahnl, Chethan Chinder Chandrappa, Celso M. de Melo, Suya You, Stefano Soatto, Alex Wong, Achuta Kadambi

ECCV 2022 · Image generation › Image restoration · approach: Dehazing and deraining · 60 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-robust contrastive feature loss (ablation)deraining network · GT-RAIN test (overall)PSNR: 21.82 -> 22.53Improved
add the rain-robust contrastive feature loss (ablation)deraining network · GT-RAIN test (overall)SSIM: 0.7148 -> 0.7304Improved
add the rain-robust loss to raise rainy/clean latent correlation (ablation)deraining network · GT-RAIN test (overall)normalized latent correlation: .85 ± .10 -> .95 ± .03Improved
retrain on GT-RAIN instead of original synthetic/semi-real dataRCDNet · GT-RAIN test (overall)PSNR: 20.26 -> 20.94Improved
retrain on GT-RAIN instead of original synthetic dataEDR · GT-RAIN test (overall)PSNR: 19.81 -> 21.44Improved
retrain on GT-RAIN instead of original synthetic dataMPRNet · GT-RAIN test (overall)PSNR: 20.09 -> 21.56Improved
retrain on GT-RAIN: SSIM of RCDNetRCDNet · GT-RAIN test (overall)SSIM: 0.6881 -> 0.7091Improved
fine-tune on GT-RAIN after synthetic training instead of original weightsMPRNet · GT-RAIN test (overall)PSNR: 20.09 -> 22.24Improved
fine-tune on GT-RAIN after synthetic training instead of training only on GT-RAINMPRNet · GT-RAIN test (overall)PSNR: 21.56 -> 22.24Improved
pretrain on synthetic data then fine-tune on GT-RAINOurs · GT-RAIN test (overall)PSNR: 22.53 -> 22.68Improved
train on GT-RAIN instead of other datasetsOurs vs MPRNet · GT-RAIN test (dense rain accumulation)PSNR: 3.40 dBImproved
train on real paired GT-RAIN rather than multiple synthetic datasetsOurs vs MPRNet · GT-RAIN test (overall)PSNR: 2.44 dBImproved

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