AI Paper Atlas

Research map / Image generation / Super-resolution

Cross-Attention in Coupled Unmixing Nets for Unsupervised Hyperspectral Super-Resolution

Jing Yao, Danfeng Hong, Jocelyn Chanussot, Deyu Meng, Xiaoxiang Zhu, Zongben Xu

ECCV 2020 · Image generation › Super-resolution · approach: Efficient lightweight super-resolution · 181 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 softmax with clamp function at encoder end (ablation)CUCaNet · Pavia UniversityPSNR (dB): 34.25 -> 35.67Improved
add spatial-spectral consistency module (ablation)CUCaNet + clamp · Pavia UniversityPSNR (dB): 35.67 -> 36.55Improved
add cross-attention module (ablation)CUCaNet + clamp · Pavia UniversityPSNR (dB): 35.67 -> 36.49Improved
add cross-attention to SSC model (ablation)CUCaNet + clamp + SSC · Pavia UniversityPSNR (dB): 36.55 -> 37.22Improved
add SSC to cross-attention model (ablation)CUCaNet + clamp + CA · Pavia UniversitySAM: 4.63 -> 4.43Improved
add cross-attention to SSC model (ablation)CUCaNet + clamp + SSC · Pavia UniversitySAM: 4.76 -> 4.43Improved
use unsupervised coupled-autoencoder network instead of CNMF (ablation)CUCaNet baseline · Pavia UniversityPSNR (dB): 32.73 -> 34.25Improved
use CUCaNet instead of supervised MHFnetCUCaNet · CAVEPSNR (dB): 37.30 -> 37.51Improved
use CUCaNet instead of supervised MHFnetCUCaNet · ChikuseiPSNR (dB): 43.71 -> 42.70Hurt
use CUCaNet instead of supervised MHFnetCUCaNet · Pavia UniversitySAM: 5.15 -> 4.43Improved

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