A deep learning pre-trained model based on the Enhanced Super Resolution Generative Adversarial Network (ESRGAN) is used to obtain a Super-Resolution (SR) image from a Low-Resolution (LR) Widefield microscopy image. The application of an ESRGAN-based model on Stochastic Optical Reconstruction Microscopy (STORM) images aims to overcome the complex sample staining process and the relatively long acquisition times for a single image. The DL models and the weights included in this repository are the specific models used to generate the experimental results reported in the paper by Lossano, S., Capaccioli, S., Cella Zanacchi, F., Da Pozzo, E., Del Debbio, F., Evelina Fantacci, M., Lizzi, F., Magrassi, R., Noferi, B., Pisignano, D., Scapicchio, C., & Retico, A. (2025). Generative super-resolution AI accelerates nanoscale analysis of cells. Machine Learning: Science and Technology, 6(2), 025001. https://doi.org/10.1088/2632-2153/adc3e9