Curriculum Learning for ab initio Deep Learned Refractive Optics

arXiv:2302.01089v3 Announce Type: replace-cross
Abstract: Deep optical optimization has recently emerged as a new paradigm for designing computational imaging systems using only the output image as the objective. However, it has been limited to either simple optical systems consisting of a single element such as a diffractive optical element (DOE) or metalens, or the fine-tuning of compound lenses from good initial designs. Here we present a DeepLens design method based on curriculum learning, which is able to learn optical designs …

abstract arxiv computational cs.cv cs.lg curriculum curriculum learning designing eess.iv element however image imaging new paradigm optical optics optimization paradigm physics.optics simple systems type