CNN-based Image Super-Resolution and Deblurring
Bianli Du, Xiaokang Ren, Jie Ren · 2019
Image super-resolution and deblurring are two highly ill-posed problems that are usually dealt separately. However, real-world images are often low-resolution and have complex blurring. In this paper, non-uniform motion blur and super-resolution task are combined to reconstruct a clear high-resolution image directly from the blurred low-resolution input. We propose a two-branch network based on convolutional neural network(CNN) to reconstruct images, which mainly include super-resolution module and deblurring module. In addition, we use novel loss functions to generate more realistic images. Experimental results show that the method proposed in this paper can reconstruct sharper high-resolution images than other state-of-the-art algorithms, and the proposed model is lightweight, requiring a low computational cost.