Conditional neural holography: a distance-adaptive CGH generator.

Area of Science:

• Optics and Photonics
• Computer Vision
• Artificial Intelligence

Background:

• Convolutional neural networks (CNNs) offer a balance between speed and accuracy in synthesizing computer-generated holograms (CGHs).
• Existing CNN methods for CGH synthesis lack the ability to specify hologram propagation distance, limiting their applicability.

Purpose of the Study:

• To develop a distance-adaptive CGH generator capable of specifying both the target image and propagation distance.
• To overcome the limitations of fixed-distance CNN-based CGH synthesis methods.

Main Methods:

• A novel distance-adaptive CGH generator was developed.
• The model incorporates a zone plate encoder stage and an augmented HoloNet stage.
• The system allows for the specification of target images and desired propagation distances.

Main Results:

• The developed model achieves comparable performance to prior fixed-distance CNN methods.
• The generator successfully synthesizes CGHs with specified propagation distances.
• The method demonstrates the accuracy and speed required for practical holographic applications.

Conclusions:

• The distance-adaptive CGH generator offers enhanced flexibility for holographic display technologies.
• This approach addresses the critical need for distance control in CNN-based hologram synthesis.
• The model provides a viable solution for practical, high-speed, and accurate CGH generation.