Physics-aware cross-domain fusion aids learning-driven computer-generated holography

作     者：GANZHANGQIN YUAN MI ZHOU FEI LIU MU KU CHEN KUI JIANG YIFAN PENG ZIHAN GENG GANZHANGQIN YUAN;MI ZHOU;FEI LIU;MU KU CHEN;KUI JIANG;YIFAN PENG;ZIHAN GENG

作者机构：Institute of Data and InformationTsinghua Shenzhen International Graduate SchoolTsinghua UniversityShenzhen 518071China School of Optoelectronic EngineeringXidian UniversityXi’an 710071China Department of Electrical EngineeringCity University of Hong KongHong Kong SARChina School of Computer Science and TechnologyHarbin Institute of TechnologyHarbin 150001China Department of Electrical and Electronic EngineeringThe University of Hong KongHong Kong SARChina Department of Computer ScienceThe University of Hong KongHong Kong SARChina 

出 版 物：《Photonics Research》 (光子学研究(英文版))

年 卷 期：2024年第12卷第12期

页      面：2747-2756页

核心收录：

学科分类：070207[理学-光学] 07[理学] 08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 0803[工学-光学工程] 0702[理学-物理学] 

基　　金：National Natural Science Foundation of China(62305184) Basic and Applied Basic Research Foundation of Guangdong Province(2023A1515012932) Science,Technology and Innovation Commission of Shenzhen Municipality(WDZC20220818100259004) Research Grants Council of the Hong Kong Special Administrative Region,China(C5031-22G,CityU11300123,CityU11310522) Guangdong Provincial Department of Science and Technology(2020B1515120073) City University of Hong Kong(9610628) Research Grants Council of Hong Kong(ECS 27212822) 

主　　题：computer cross holography 

摘      要：The rapid advancement of computer-generated holography has bridged deep learning with traditional optical principles in recent ***,a critical challenge in this evolution is the efficient and accurate conversion from the amplitude to phase domain for high-quality phase-only hologram(POH)*** computational models often struggle to address the inherent complexities of optical phenomena,compromising the conversion *** this study,we present the cross-domain fusion network(CDFN),an architecture designed to tackle the complexities involved in POH *** CDFN employs a multi-stage(MS)mechanism to progressively learn the translation from amplitude to phase domain,complemented by the deep supervision(DS)strategy of middle features to enhance task-relevant feature learning from the initial ***,we propose an infinite phase mapper(IPM),a phase-mapping function that circumvents the limitations of conventional activation functions and encapsulates the physical essence of *** simulations,our proposed method successfully reconstructs high-quality 2K color images from the DIV2K dataset,achieving an average PSNR of 31.68 dB and SSIM of ***,we realize high-quality color image reconstruction in optical *** experimental results highlight the computational intelligence and optical fidelity achieved by our proposed physics-aware cross-domain fusion.