AI Paper Atlas

Research map / Image generation / Image restoration

Stochastic Window Transformer for Image Restoration

Jie Xiao, Xueyang Fu, Feng Wu, Zheng-Jun Zha

NeurIPS 2022 · Image generation › Image restoration · approach: Efficient restoration architectures · 1 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
train with stochastic window instead of fixed window (test fixed) (ablation)Stoformer · SPA-Data derainingPSNR: 47.80 -> 48.85Improved
add layer expectation propagation at test on top of stochastic training (ablation)Stoformer · SPA-Data derainingPSNR: 48.85 -> 48.97Improved
apply layer expectation propagation at test without stochastic training (ablation)Stoformer♢♢ · SPA-Data derainingPSNR: 47.80 -> 46.95Hurt
train with stochastic window instead of fixed window (test fixed) (ablation)Stoformer · GoPro deblurringPSNR: 32.80 -> 33.17Improved
add layer expectation propagation at test on top of stochastic training (ablation)Stoformer · GoPro deblurringPSNR: 33.17 -> 33.24Improved
apply layer expectation propagation at test without stochastic training (ablation)Stoformer♢♢ · GoPro deblurringPSNR: 32.80 -> 31.62Hurt
apply layer expectation propagation at test without stochastic training (ablation)Stoformer♢♢ · HIDE deblurringPSNR: 30.73 -> 29.94Hurt
stochastic window training and expectation propagation vs fixed window (ablation)Stoformer · CBSD68 color denoising sigma=25PSNR: 31.73 -> 32.47Improved
stochastic window training and expectation propagation vs fixed window (ablation)Stoformer · Urban100 grayscale denoising sigma=50PSNR: 27.90 -> 28.72Improved
stochastic window training and expectation propagation vs fixed window (ablation)Stoformer · Set12 grayscale denoising sigma=50PSNR: 27.96 -> 28.46Improved
stochastic Stoformer vs RestormerStoformer†† · Kodak24 color denoising sigma=25PSNR: 33.02 -> 33.12No clear effect
stochastic Stoformer vs MPRNetStoformer†† · HIDE deblurringPSNR: 30.96 -> 30.99No clear effect

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.

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