AI Paper Atlas

Research map / Image generation / Super-resolution

Image Super-Resolution with Deep Dictionary

Shunta Maeda

ECCV 2022 · Image generation › Super-resolution · approach: Real-world super-resolution

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 batch normalization (ablation)SRDD-64 · Set14 x4PSNR: 28.49 -> 28.54Improved
add bottleneck blocks and D_H code injection (ablation)SRDD-64 · Set14 x4PSNR: 28.48 -> 28.54Improved
add compensation (ablation)SRDD-64 · Set14 x4PSNR: 28.51 -> 28.54Improved
increase atoms from 64 to 128 (ablation)SRDD · Set14 x4 in-domainPSNR: 28.54 -> 28.65Improved
increase atoms from 64 to 128SRDD · Real-world historical imagesNIQE: 5.877 -> 5.896Hurt
use SRDD-64 instead of CARN on OpenCV bicubic downsampled inputSRDD-64 · Set14 x4PSNR: 21.17 -> 21.52Improved
use SRDD-64 instead of CARN on OpenCV bilinear downsampled inputSRDD-64 · Set14 x4PSNR: 22.76 -> 23.13Improved
use SRDD-64 instead of CARN on OpenCV area downsampled inputSRDD-64 · Set14 x4PSNR: 26.74 -> 27.05Improved
use SRDD-64 instead of CARNSRDD-64 · Real-world historical imagesNIQE: 5.921 -> 5.877Improved
use ReLU instead of softmax for coefficients (ablation)SRDD · training stabilitystabilityHurt
use SRDD-64 instead of CARN in-domainSRDD-64 · Set14 x4 (Matlab bicubic)PSNR: 28.60 vs 28.54No clear effect

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