… We develop a robust image super-resolution network with the proposed frequency mask …
could be an image or a feature map extracted by super-resolution networks. First, the input …

… INTRODUCTION SINGLE image super-resolution is usually cast as an … fully convolutional
neural network to directly learn the mapping from the input LR image and the output HR image. …

… )-based face image super-resolution approaches are able to … a novel SR-based face image
superresolution approach that … aim at learning the direct mapping function from the LR space …

… In this paper, we introduce a robust GAN for image superresolution to be able to mitigate the
… to produce 200 feature maps as the first convolutional layer and 300 feature maps as the …

… mainly map the whole LR image to the whole HR image, while the algorithms based on the
local structure map … Her research interests include image super resolution and related vision …

… resolution images) we can learn set of filters (ie a mapping) … to be used in order to obtain a
reliable estimate. In fact, we … Single Image Super-Resolution In this subsection we compare …

… It is commonly recognized that these tasks have strong … up sample the LR feature maps in
an approach, which is also … an ablation study on the image super-resolution task. Table I lists …

… in the field of single image super-resolution (SR) compared with … LR images, we propose a
multi-scale feature mapping scheme … Robust single image super-resolution via deep networks …

… Learning-based single image super-resolution (SISR) … of a data term and a prior term under
the MAP framework. … Therefore, it is necessary to further analyze the kernel robustness of …

… Single image Super-Resolution (SISR) aims to generate a … interpolated LR images, failing
to establish an end-to-end mapping from … the effectiveness and robustness of our RDN model. …