A deep learning approach to image analysis of M87⋆

ABSTRACT Deep learning-based computer vision models have demonstrated superior performance on a wide variety of image-related tasks, including image classification and object recognition. This study explores the potential of deep learning techniques for analysing images of M87$^{\star }$, using both simulated and real observational data. We investigate the performance of various neural network architectures for parameter inference, multitask learning, and super-resolution imaging. This work serves as a preliminary step towards developing robust deep learning methods for interpreting black hole imagery and advancing our understanding of black hole physics.