83‐3: Inverse Design of Liquid Crystal Phase Modulators for 2D/3D Switchable Display Based on Deep Learning

Abstract

The fringe field effect in liquid crystal (LC) devices can cause pixel crosstalk issues. Therefore, LC devices with multiple electrode structures require multiple calibrations to achieve an approximate distribution of the target phase modulation. At the same time, optical inverse design based on data‐driven algorithms is also a highly popular topic. This article introduces a deep‐learning approach to realize inverse phase modulation in LC devices. By adopting the deep learning method for inverse phase modulation in LC devices, can effectively alleviate the problems caused by fringe field effects and achieve more precise and accurate phase modulation distribution.