Towards a computer vision particle flow

Abstract In High Energy Physics experiments Particle Flow (PFlow) algorithms are designed to provide an optimal reconstruction of the nature and kinematic properties particles produced within detector acceptance during collisions. At heart PFlow is ability distinguish calorimeter energy deposits neutral from those charged particles, using complementary measurements particle tracking devices, a superior measurement content kinematics. this paper, computer vision approach fundamental aspect algorithms, based on images, proposed. A comparative study state art deep learning techniques performed. significantly improved obtained in context large overlaps with particles. Calorimeter images augmented finer granularity also super-resolution techniques.