Neural network based 3D tracking with a graphene transparent focal stack imaging system

Abstract Recent years have seen the rapid growth of new approaches to optical imaging, with an emphasis on extracting three-dimensional (3D) information from what is normally a two-dimensional (2D) image capture. Perhaps most importantly, rise computational imaging enables both physical layouts components and algorithms be implemented. This paper concerns convergence two advances: development transparent focal stack system using graphene photodetector arrays, expansion capabilities machine learning including powerful neural networks. demonstrates 3D tracking point-like objects multilayer feedforward networks extension positions multi-point objects. Computer simulations further demonstrate how this can track extended in 3D, highlighting promise combining nanophotonic devices, designs, for frontiers imaging.