AI Paper Atlas

Research map / Image generation / Super-resolution

Learning Frequency-Aware Dynamic Network for Efficient Super-Resolution

Wenbin Xie, Dehua Song, Chang Xu, Chunjing Xu, Hui Zhang, Yunhe Wang

ICCV 2021 · Image generation › Super-resolution · approach: CNN and transformer SR architectures · 83 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
replace EDSR with frequency-aware dynamic network (FAD-EDSR)EDSR · Set5 x2FLOPs (G): 1338.8 -> 408.4Improved
replace EDSR with FAD-EDSREDSR · Set5 x2PSNR (dB): 38.11 -> 38.21Improved
replace RCAN with FAD-RCANRCAN · Urban100 x4FLOPs (G): 1452.5 -> 891.2Improved
use FAD-EDSR instead of AdaEDSRAdaEDSR · Urban100 x4PSNR (dB): 26.58 -> 26.70Improved
deploy FAD-EDSR on single core CPUEDSR · CPU latency (single core)latency (s): 39.60s -> 25.44sImproved
deploy FAD-EDSR on double core CPUEDSR · CPU latency (double core)latency (s): 20.08s -> 13.74sImproved
use DCT frequency mask instead of random masks per image (ablation)FADN (EDSR-baseline) · Urban100 x2PSNR (dB): 31.17 -> 31.72Improved
use random masks per block instead of per image (ablation)FADN (EDSR-baseline) · Urban100 x2PSNR (dB): 31.17 -> 31.47Improved
use learnable mask instead of DCT mask (no guidance) (ablation)FADN (EDSR-baseline) · Urban100 x2PSNR (dB): 31.72 -> 31.81Improved
add DCT guidance to learnable mask (ablation)FADN (EDSR-baseline) · Urban100 x2PSNR (dB): 31.81 -> 31.90Improved
increase branches from 2 to 3 (ablation)FADN (EDSR-baseline) · Urban100 x2PSNR (dB): 31.72 -> 31.85Improved
increase branches from 3 to 4 (ablation)FADN (EDSR-baseline) · Urban100 x2PSNR (dB): 31.85 -> 31.88No 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