AI Paper Atlas

Research map / Image generation / Image restoration

Blind Image Decomposition

Junlin Han, Weihao Li, Pengfei Fang, Chunyi Sun, Jie Hong, Mohammad Ali Armin, Lars Petersson, Hongdong Li

ECCV 2022 · Image generation › Image restoration · approach: Dehazing and deraining · 55 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 multi-scale encoder (vs single-scale) (ablation)BIDeN · Task I (N=4)Fruit PSNR: 17.26 -> 18.19Improved
add adversarial loss (ablation)BIDeN · Task I (N=4)Acc (AB): 0.566 -> 0.716Improved
add perceptual loss (ablation)BIDeN · Task I (N=4)Fruit PSNR: 16.67 -> 18.19Improved
add L1/L2 loss (ablation)BIDeN · Task I (N=4)Fruit PSNR: 15.56 -> 18.19Improved
add binary cross-entropy loss for source prediction (ablation)BIDeN · Task I (N=4)Acc (AB): 0.0 -> 0.716Improved
move source prediction branch from discriminator into generator (ablation)BIDeN · Task I (N=4)Acc (AB): 0.716 -> 0.634Hurt
share weights between prediction and separation discriminator branches (ablation)BIDeN · Task I (N=4)Acc (AB): 0.520 -> 0.716Improved
add zeroed loss on unused heads (ablation)BIDeN · Task I (N=4)Fruit PSNR: 18.19 -> 15.68Hurt
train on BID setting rather than conventional decompositionBIDeN · Task II.B real snowBRISQUE: 22.64 -> 22.40No clear effect
train on BID setting rather than conventional decompositionMPRNet · Task II.B real raindropBRISQUE: 29.42 -> 25.86Improved
increase max number of source components N from 2 to 8BIDeN · Task I Fruit PSNRPSNR: 20.07 -> 16.49Hurt
increase max number of source components N from 2 to 8BIDeN · Task I Acc (All)Acc: 0.957 -> 0.278Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

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

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