We propose a deep hologram converter based on deep learning to convert low-precision holograms into middle-precision holograms. The low-precision holograms were calculated using a shorter bit width. It can increase the amount of data packing for single instruction/multiple data in the software approach and the number of calculation circuits in the hardware approach. One small and one large deep neural network (DNN) are investigated. The large DNN exhibited better image quality, whereas the smaller DNN exhibited a faster inference time. Although the study demonstrated the effectiveness of point-cloud hologram calculations, this scheme could be extended to various other hologram calculation algorithms.
Shiomi, H, Blinder, D, Birnbaum, T, Inoue, Y, Wang, F, Ito, T, Kakue, T, Schelkens, P & Shimobaba, T 2023, 'Deep hologram converter from low-precision to middle-precision holograms', Appl. Optics, vol. 62, no. 7, pp. 1723-1729. https://doi.org/10.1364/AO.482434
Shiomi, H., Blinder, D., Birnbaum, T., Inoue, Y., Wang, F., Ito, T., Kakue, T., Schelkens, P., & Shimobaba, T. (2023). Deep hologram converter from low-precision to middle-precision holograms. Appl. Optics, 62(7), 1723-1729. https://doi.org/10.1364/AO.482434
@article{a3626785e49f49c781219c56747c32ec,
title = "Deep hologram converter from low-precision to middle-precision holograms",
abstract = "We propose a deep hologram converter based on deep learning to convert low-precision holograms into middle-precision holograms. The low-precision holograms were calculated using a shorter bit width. It can increase the amount of data packing for single instruction/multiple data in the software approach and the number of calculation circuits in the hardware approach. One small and one large deep neural network (DNN) are investigated. The large DNN exhibited better image quality, whereas the smaller DNN exhibited a faster inference time. Although the study demonstrated the effectiveness of point-cloud hologram calculations, this scheme could be extended to various other hologram calculation algorithms.",
author = "Harutaka Shiomi and David Blinder and Tobias Birnbaum and Yota Inoue and Fan Wang and Tomoyoshi Ito and Takashi Kakue and Peter Schelkens and Tomoyoshi Shimobaba",
note = "Funding Information: Funding. Japan Society for the Promotion of Science (19H01097, 22H03607, JPJSBP120202302); IAAR Research Support Program, Chiba University; The Joint JSPS–FWO Scientific Cooperation Program (VS07820N); The FWO Junior and Senior Postdoctoral Fellowships (12ZQ220N, 12ZQ223N); Korea Institute for Advancement of Technology (P146500071). Publisher Copyright: {\textcopyright} 2023 Optica Publishing Group. Copyright: Copyright 2023 Elsevier B.V., All rights reserved.",
year = "2023",
month = mar,
day = "1",
doi = "10.1364/AO.482434",
language = "English",
volume = "62",
pages = "1723--1729",
journal = "Appl. Optics",
issn = "1559-128X",
publisher = "The Optical Society",
number = "7",
}