.png){kind=icon}
Return
Continuous Remote Sensing Image Super-Resolution Based on Context Interaction in Implicit Function Space
.png){kind=icon}
OA
AI
DOI：10.1109/TGRS.2023.3272473.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Despite its fruitful applications in remote sensing, image super-resolution (SR) is troublesome to train and deploy as it handles different resolution magnifications with separate models. Accordingly, we propose a highly applicable SR framework called FunSR, which settles different magnifications with a unified model by exploiting context interaction within implicit function space. FunSR composes a functional representor, a functional interactor, and a functional parser. Specifically, the representor transforms the low-resolution image from Euclidean space to multiscale pixelwise function maps; the interactor enables pixelwise function expression with global dependencies; and the parser, which is parameterized by the interactor's output, converts the discrete coordinates with additional attributes to RGB values. Extensive experimental results demonstrate that FunSR reports the state-of-the-art performance on both fixed- and continuous-magnification settings; meanwhile, it provides many friendly applications due to its unified nature. Our code is available at https://github.com/KyanChen/FunSR.
Keywords:
Remote sensing
Superresolution
Spatial resolution
Semantics
Image edge detection
Generative adversarial networks
Deep learning
Continuous image representation
function space
implicit neural network
remote sensing images
super-resolution (SR)
Journal
IF:
8.6
Papers:
2.2W
Citations:
10.7W
Organization
Cited Papers
.png){kind=icon}
ShareSaveImage super-resolution using progressive generative adversarial networks for medical image analysis
PREAI
ShareSaveShareSaveShareSave
View more