Learning-based image coding has shown promising results during recent years. Unlike the traditional approaches to image compression, learning-based codecs exploit deep neural networks for reducing dimensionality of the input at the stage where a linear transform would be typically applied previously. The signal representation after this stage, called latent space, carries the information in such a way that it can be interpreted by other deep neural networks without the need of decoding it. One of the tasks that can benefit from the above-mentioned possibility is super resolution. In this paper, we explore the possibilities and propose an approach for super resolution that is applied in the latent space. We focus on the fixed compression model, where the encoder part of the network is frozen and an enhanced decoder is learned. Additionally, we assess the performance of the proposed approach.