In this paper, we propose a foveated super resolution (SR) algorithm to create high resolution images from low resolution inputs for virtual reality head mounted displays. The proposed SR algorithm integrated in the foveation‐based driving technology covers the small area around the foveation point that requires high rendering quality, substantially reducing the overall computational complexity. We employ cropping as well as progressive and recursive framework and reduce the number of neurons by placing the deconvolutional layer after convolutional layers.