Image Generation with Super Resolution
- Image super-resolution (SR) is a task of enhancing the resolution of images
- With the help of fully convolutional networks, finer and realistic high-resolution images can be generated from low-resolution images
- perceptual loss which consists of an adversarial loss and a content loss
- With the perceptron loss, the SRGAN is able to recover photo-realistic textures from heavily downsampled images and show significant gains in perceptual quality.
- The networks for image super-resolution task are able to learn the semantic features of images