AI Paper Atlas

Research map / Image generation / Super-resolution

Single Pair Cross-Modality Super Resolution

Guy Shacht, Dov Danon, Sharon Fogel, Daniel Cohen-Or

CVPR 2021 · Image generation › Super-resolution · approach: Efficient lightweight super-resolution · 14 citations (OpenAlex)

Check 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
Use single-pair CMSR instead of VTSRGANCMSR · ULB17-VTPSNR: 27.988 -> 29.928Improved
Use single-pair CMSR instead of ZSSRCMSR · ULB17-VTSSIM: 0.8567 -> 0.882Improved
Use single-pair CMSR instead of VTSRCNNCMSR · Shuffled-MiddleburyPSNR: 28.189 -> 28.652Improved
Use single-pair CMSR instead of VTSRGANCMSR · Shuffled-MiddleburySSIM: 0.9547 -> 0.9341Hurt
Use single-pair CMSR instead of VTSRGANCMSR · EPFL NIRPSNR: 30.665 -> 31.201Improved
Use single-pair CMSR instead of VTSRGANCMSR · EPFL NIRSSIM: 0.9005 -> 0.9200Improved
Add the affine transformation layer (ablation)CMSR coarse-to-fine deformation framework · Weakly aligned visual-thermal pairPSNRImproved
Add the CPAB transformation layer (ablation)CMSR coarse-to-fine deformation framework · Weakly aligned visual-thermal pairPSNRImproved
Add the TPS transformation layer (ablation)CMSR coarse-to-fine deformation framework · Weakly aligned visual-thermal pairPSNRImproved
Alternate upsampling-based training with downsampling-based training instead of using only downsampling-based training (ablation)CMSR · Weakly aligned visual-thermal pairPSNRImproved
Set the upsampling-based training probability to p=0.3 (ablation)CMSR alternating-scales training · Weakly aligned visual-thermal pairPSNR: 32.476 dBImproved
Increase the upsampling-based training probability beyond p=0.3 (ablation)CMSR alternating-scales training · Weakly aligned visual-thermal pairPSNRHurt

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