Deep-Learning-Based Dynamic Range Compression for 3D Scene Hologram
 
Deep-Learning-Based Dynamic Range Compression for 3D Scene Hologram ■
 
Tomoyoshi Shimobaba, Tomoyoshi Shimobaba, Tomoyoshi Shimobaba, David Blinder, David Blinder, David Blinder, Peter Schelkens, Peter Schelkens, Peter Schelkens, Yota Yamamoto, Yota Yamamoto, Yota Yamamoto, Ikuo Hoshi, Ikuo Hoshi, Ikuo Hoshi, Atsushi Shiraki, Atsushi Shiraki, Atsushi Shiraki, Takashi Kakue, Takashi Kakue, Takashi Kakue, Tomoyoshi Ito, Tomoyoshi Ito, Tomoyoshi Ito
 
Abstract ■

This study proposes a dynamic-range compression for digital holograms generated from three-dimensional scenes using deep neural network (DNN). This method uses an error diffusion algorithm to binarize holograms with an 8-bit gradation; moreover, the DNN predicts the original gradation holograms from binary holograms. This method{\textquoteright}s performance exceeds that of JPEG 2000 and high-efficiency video coding.