Cross-Scale KNN Image Transformer for Image Restoration

Hunsang Lee, Hyesong Choi, Kwanghoon Sohn, Dongbo Min

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


Numerous image restoration approaches have been proposed based on attention mechanism, achieving superior performance to convolutional neural networks (CNNs) based counterparts. However, they do not leverage the attention model in a form fully suited to the image restoration tasks. In this paper, we propose an image restoration network with a novel attention mechanism, called cross-scale k -NN image Transformer (CS-KiT), that effectively considers several factors such as locality, non-locality, and cross-scale aggregation, which are essential to image restoration. To achieve locality and non-locality, the CS-KiT builds k -nearest neighbor relation of local patches and aggregates similar patches through local attention. To induce cross-scale aggregation, we ensure that each local patch embraces different scale information with scale-aware patch embedding (SPE) which predicts an input patch scale through a combination of multi-scale convolution branches. We show the effectiveness of the CS-KiT with experimental results, outperforming state-of-the-art restoration approaches on image denoising, deblurring, and deraining benchmarks.

Original languageEnglish
Pages (from-to)13013-13027
Number of pages15
JournalIEEE Access
StatePublished - 2023

Bibliographical note

Publisher Copyright:
© 2013 IEEE.


  • Image restoration
  • deblurring
  • denoising
  • deraining
  • k-nn search
  • low-level vision
  • self-attention
  • transformer
  • transformer


Dive into the research topics of 'Cross-Scale KNN Image Transformer for Image Restoration'. Together they form a unique fingerprint.

Cite this