/super_resolution

Enhancing the resolution of images using SRCNN

Primary LanguageJupyter NotebookMIT LicenseMIT

super_resolution

Enhancing the resolution of images using SRCNN i.e recover a high resolution image from a low resolution input. Keras is being used for making this super-resolution convolution neural network(SRCNN) model.

The result is shown below:

srcnn

  • Original Paper: "Image-Super-Resolution Using Deep Convolutional Networks" by Chao Dong et al. 2014. link to the paper.
    The pretrained weights are taken from Here