articleIEEE Transactions on Image ProcessingJan 1, 2024Closed access

TTST: A Top- k Token Selective Transformer for Remote Sensing Image Super-Resolution

Wuhan University · Harbin Institute of Technology · +2 more institutions

PubMed
Indexed incrossrefpubmed

Abstract

Transformer-based method has demonstrated promising performance in image super-resolution tasks, due to its long-range and global aggregation capability. However, the existing Transformer brings two critical challenges for applying it in large-area earth observation scenes: (1) redundant token representation due to most irrelevant tokens; (2) single-scale representation which ignores scale correlation modeling of similar ground observation targets. To this end, this paper proposes to adaptively eliminate the interference of irreverent tokens for a more compact self-attention calculation. Specifically, we devise a Residual Token Selective Group (RTSG) to grasp the most crucial token by dynamically selecting the…

Citation impact

312
total citations
FWCI
70.01
Percentile
100%
References
72
Citations per year

Authors

6

Topics & keywords

Keywords
  • Security token
  • Computer science
  • Transformer
  • Artificial intelligence
  • Pattern recognition (psychology)
  • Voltage
  • Engineering
  • Computer network
No related works found for this paper.

Funding