logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

PTSR: A patch-based translator model for image super-resolution

delete2025-05-26
delete0
PRE
AI
S
S. Bahinipati
S
Shiv Ram Dubey
S
Satish Kumar
DOI：10.1016/j.patrec.2025.04.020delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Fully transformer-based model eliminates convolutions in both generator and discriminator. • Novel PTSR module captures local textures and global context for SR tasks. • Separates features and position encoding for better spatial consistency. • Outperforms BSRGAN, Real-ESRGAN, IPT, and FlowIE on multiple benchmarks.

Journal

Pattern Recognition Letters cover
Pattern Recognition Letters
IF:
3.3
Papers:
7.9K
Citations:
1.6W

Organization

No organization information available