Neural Network Assisted End‐To‐End Design for Broadband High Diffraction Efficiency Full‐Color Meta‐Holograms
Year: 2025
Type: article
Abstract: Metasurface holography has emerged as a revolutionary field with enormous potential in augmented reality (AR), virtual reality (VR), as well as 3D imaging. Despite significant advances, metasurfaces in the visible band are hindered by critical limitations, like low diffraction efficiency and narrow‐bandwidth, which severely constrain their practical applications. In this study, a low‐loss dielectric meta‐atoms designed is proposed by a neural network‐assisted end‐to‐end method, achieving an extraordinarily high and uniform diffraction efficiency over the visible band. The metasurface, constructed by uniformly‐distributed single‐sized meta‐atoms in a periodic arrangement, can achieve full color phase‐only holograms. In addition, this meta‐atom can also enable potential functions of metasurface over the entire visible band, like metalens and vortex generators, while maintaining high diffraction efficiency. This work represents a profound shift in the field of high‐efficiency, full‐color holographic displays, providing unprecedented opportunities for AR/VR and other applications.
Source: Laser & Photonics Review
Authors: Huan Yuan, Zheqiang Zhong, Bin Zhang
Institution: Sichuan University
Language: English
Field: Materials Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4413240182