AI Paper Atlas

Research map / Image generation / Image restoration

Asymmetric Mask Scheme for Self-Supervised Real Image Denoising

Xiangyu Liao, Tianheng Zheng, Jiayu Zhong, Pingping Zhang, Chao Ren

ECCV 2024 · Image generation › Image restoration · approach: Self-supervised image denoising · 14 citations (OpenAlex)

Read the paper

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 random replacement refinement at inference (ablation)AMSNet-P · SIDD ValidationPSNR (dB): 36.71 -> 37.93Improved
add smoothness loss fine-tuning L_t (ablation)AMSNet-B · SIDD ValidationPSNR (dB): 36.60 -> 36.71Improved
use mask-based training instead of blind-spot with unrestricted receptive field (ablation)AMSNet vs AP-BSN · SIDD ValidationPSNR (dB): 20.91 -> 37.11Improved
use Restormer instead of DnCNN as denoiser in AMSNet (ablation)AMSNet · SIDD ValidationPSNR (dB): 36.93 -> 37.93Improved
set mask ratio per branch to about 50%AMSNet · SIDD ValidationPSNR (dB)Improved
use AMSNet-B without refinement versus AP-BSN+R3AMSNet-B · SIDD BenchmarkPSNR (dB): 36.60 vs 36.91No clear effect
use AMSNet-P-E instead of best prior self-supervised methodsAMSNet-P-E · SIDD BenchmarkPSNR (dB): 37.41 -> 37.87 vs BNN-LANImproved
use AMSNet-P-E on PolyUAMSNet-P-E · PolyUPSNR (dB): 37.13 -> 37.92 vs BNN-LANImproved

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