Practical Holography XXXIX: Displays, Materials, and Applications

We describe the requirements for augmented reality (AR) displays, and describe how a laser-illuminated holographic display technology could satisfy them. We present the basic operating principles of such a display and, by using a simple first-order analysis, we derive initial specifications for a holographic AR display that could be realised by Swave’s HXR modulator. Finally, we present initial results of a full-colour holographic display based on spatial colour.

In this study, we investigate the removal of zero-order diffraction (ZOD) in computer-generated holography (CGH) using a destructive interference approach. Our method addresses the common issue of unmodulated light, which manifests as a bright spot at the center of virtual images, by generating a counter-phase light pattern. We demonstrate that our approach can achieve up to $86\pm{2}\%$ reduction in ZOD intensity across various holographic techniques, including Fourier and Fresnel transform holography. This method ensures constant alignment, does not compromise the modulation depth of Spatial Light Modulators (S.L.M.), and is computationally efficient. Our findings pave the way for more effective integration of CGH in augmented reality (AR) applications, particularly in near-eye displays (NEDs), by providing clearer and more accurate virtual images.

A Depth-Fused 3-D Display (DFD) is known as a three-dimensional display method using a visual perception fusion of two different depth 2-D images, where continuous depth structure can be recognized [1]. Transparent displays are required to display the DFD 3-D images. A see through aerial display system using a Dihedral Corner Reflector Array (DCRA) and holographic mirrors has high quality transparency characteristics by the volume holographic HOEs which have the large wavelength and angler selectivity. By using the characteristics of Bragg diffractions functionally, this system can suppress the DCRA’s virtual images and can compensate the HOE’s color dispersion, simultaneously and effectively [3]. We applied this see through aerial display method to the DFD projection system to realize a see through three-dimensional aerial display. 1) Vision Research 44 (2004) 785–793. 2) Proc. Fall meeting. Jpn. Soc. Appl. Phys. (2008), 19a-221B-3 3) Appl. Opt. 60, 31 (2021) 9896-9905.

We investigate how to efficiently compress phase-only holograms generated from 3D scenes. To tackle this challenge, we first formulate a phase unwrapping problem for phase-only holograms. We then solve this problem by using an unsupervised deep learning approach to obtain unwrapped phase-only holograms. We finally compress the unwrapped phase-only holograms, which exhibit spatial correlations similar to those of natural images, using an image coding approach. Our method demonstrates the capability to significantly reduce the bit amount required for phase-only holograms while preserving the quality of displayed 3D scenes.

Computer-generated holography (CGH) generates holograms to reconstruct three-dimensional (3D) scenes. Traditional CGH methods decompose 3D scenes into multiple planes and simulate light propagation between them, which can be computationally demanding. We introduce a novel model that simulates light propagation from a single hologram plane to multiple planes in one pass, reducing computational complexity and enhancing the efficiency of CGH algorithms.

We propose and experimentally demonstrate a holographic technique based on interfering the optimum orthogonal communication modes connecting a source plane and a receiver volume. Computing the singular value decomposition of a coupling operator, which connects each point at the source plane to another one in the receiver volume, allows us to obtain a complete set of orthogonal eigenfunctions which represent the building blocks of our hologram. By adding these eigenfunctions, weighted by different complex coefficients akin to a Fourier series, we construct arbitrarily chosen 2D and 3D images within the output receiver volume with continuous depth, high fidelity and contrast. We envision our work to inspire new directions in digital holography, AR/VR, and wearable devices.

Recently, Augmented Reality (AR) near-eye display (NED) glasses approach to commercialization thanks to technological advancements. Most AR NED glasses use transparent optical combiners, such as diffractive waveguides and holographic optical element reflectors, to display AR images on a virtual plane. In these types of AR glasses, the overall optical systems more complicated. In this study, we investigate the see-through characteristics of a direct-view AR NED using a transparent LCD with wave optic modeling and simulation. In particular, we simulate diffraction problems caused by the pixel layout of transparent displays and light smearing at points where a bright light source is located and propose a panel layout optimization method to address these issues. Additionally, we introduce a computer-generated hologram (CGH) optimization algorithm that enhances the visibility of holographic images for viewer through the virtual plane in the direct-view AR NED with the proposed panel structure.

Holographic displays hold great promise due to their ability to accurately reproduce all depth and focal cues. This paper introduces a new method called "Wireframe Holography" for generating 3D objects composed of linear segments. By incorporating controlled astigmatism onto the foci of point cloud methods, our approach creates smoother line-based objects, reduces speckle noise, and allows for the creation of more complex 3D images with similar computational costs. Compatible with existing algorithms, our method integrates seamlessly into current systems. We demonstrate its effectiveness through various 3D wireframe structures using a phase-only spatial light modulator (SLM) within a 4f system. Our results show improved performance, highlighting its potential for future holographic displays.

In this work, we report the dependence of the average photophoretic trap rate on spherical aberration, an underexplored technique in airborne photophoretic particle levitation. Using a Revibro piezoelectrically-actuated tunable focus mirror in a 633nm system, we tested deformations from -30 to +85 nm and measured wavefronts with a Shack-Hartmann sensor. Our setup, which included a vacuum chamber and automated detection system, ensured that spherical aberration was the only variable manipulated. Our findings, supported by statistical analysis, aim to enhance the understanding and efficiency of aberration-based traps, potentially broadening their applications in particle levitation technologies.

In this paper, we propose a next-generation holographic spatial display method for mobility applications, focusing on creating holographic optical elements (HOEs) optimized for curved glass surfaces, such as those found in vehicles. Using stereographic optical techniques, we aim to develop a display that appears to float in space, providing enhanced convenience for both autonomous driving systems and drivers. This advanced display technology will enable clearer and more intuitive visual information, significantly improving the driving experience and safety. Our research paves the way for the future of in-car displays, offering innovative solutions tailored to the unique needs of the automotive industry.

Generation of light patterns through computer-generated holography is limited by spatial light modulators with finite resolution and pixel pitch, and their capability to modulate only either phase or amplitude. These limitations lead to finite far-field sampling and speckle-like noise. In this study, we consider the signal bandwidth of the propagating wavefront to understand the cause of the noise. We use it to develop a new algorithm to simultaneously generate a batch of phase-only holograms using a gradient descent. When displayed in sequence, the holograms improve the time-averaged quality and increase the resolution of the projected reconstructed image compared to alternative approaches. We verify the results both numerically and experimentally. This method will benefit precision-critical fields such as holographic photolithography.

We showcase INTERFERE, a hologram compression framework that was selected as the basis of the JPEG Pleno Holography standard. It supports view-dependent coding with simultaneous spatial and angular random access. In this work, we design a new entropy coding mechanism with division-free binary arithmetic coding and achieve class-leading rate-distortion performance, ease of random access and encoding/decoding throughputs, providing a practical solution to one of digital holography's most pressing challenges.

Recent work in optical-see-through (OST) near-eye display shows significant process in expanding the field of view (FoV) and eyebox. However, improving the image quality of computer-generated holography (CGH) still poses a major challenge in providing an immersive experience of augmented reality (AR). The difficulty in CGH design lies in bridging the gap between the physical system and the idealized wave propagation model. In this work, we incorporate a camera into the full-color 3D CGH optimization of a waveguide with two dimensional eyebox expansion. This iterative optimization process is repeated across a dataset of 3D CGH patterns to train a neural network applicable to the optimization of previously unseen CGH patterns. The proposed method is shown to outperform conventional approaches in optimizing CGH patterns for two-dimensionally expanded waveguide holography.

Form factor compactness is an essential factor of augmented reality (AR) near-eye display. To achieve this compactness, the optimal design of a single diffractive waveguide wherein in-coupler, expander, and out-coupler are monolithically integrated and able to control the RGB primary colors simultaneously in a single waveguide form factor. we present a new design of achromatic diffractive waveguide with meta-grating based in-coupler, expander, and out-coupler. The grating design algorithm is the gradient optimization algorithm based on the auto-differentiation of the Fourier modal method (FMM). By imposing additional constraints to the figure of merit (FOM) to take into account fabrication feasibility, we aimed to construct the more practical design rule.

Optical metasurfaces manipulate light at subwavelength scales, providing unprecedented control. However, their design requires sophisticated multi-scale modeling algorithms in the inverse design process. Slow but rigorous computational electromagnetic methods (CEMs) like RCWA are reserved for small scale calculations (meta-atoms) while faster scalar diffraction models are used at the large scale (metasurfaces). It is typically the small scale calculations that frustrate the design process. AI techniques, such as surrogate models using artificial neural networks like Auto-Encoders (AEs), offer a solution by approximating rigorous CEMs with similar accuracy but much faster execution. Normally, these surrogate models streamline the small scale inverse design process by constructing libraries of meta-atoms for the large scale optimization to use. We demonstrate an alternative approach that applies an AE surrogate model directly to a fast/flexible adjoint-based scalar diffraction inverse design algorithm (DNA)2. Our results will include robust multilayer, multi-wavelength, polarization manipulating metasurface holograms.

Holography, and the production of holograms, is one of the rare fields in which artists and scientists are concerned about the same field of action. The difference in processes lies in the fact that scientists will create images that strictly respond to technical requirements, while artists will aim to create sensitive images that are the expression of their thoughts. For people who are not familiar with holograms, it is not easy to make them interested in holograms through a photography or a video because they cannot convey the holographic work completely, so it is important to see a hologram directly and experience the holographic environment because this is the nature of the holographic medium. Exhibitions thus become a means of promoting this innovative art form. This paper presents work developed and results obtained in artistic holography and technical holography that combines art and science to promote public engagement in holography.

Computer-Generated Holography (CGH) offers authentic three-dimensional visualization through the full wavefront reconstruction of a scene, utilizing a coherent light source's controlled diffraction via a spatial light modulator (SLM). Existing SLMs are restricted to modulating either amplitude or phase exclusively. Consequently, specialized algorithms are essential for generating phase-only holograms. Traditional phase-retrieval techniques for CGH focus on deriving phase-only holograms from amplitude-only target images, often neglecting the phase of the target field. This study introduces an innovative hologram generation approach that employs an optimization algorithm to search for an optimal phase to be multiplied with the target image, whose backward propagation generates a hologram with an amplitude that is as uniform as possible. The resulting phase of this hologram is then the phase hologram suitable for phase-only SLMs.

Learning-based computer-generated hologram (CGH) optimization significantly achieves high-quality holographic three-dimensional images using phase-only spatial light modulators (SLMs). This paper presents a CGH synthesis framework for complex holographic image observations by finite pupil of viewer’s eye, which plays a role of optical filter. We devise a numerical wave optical model to describe observation of holographic images by human vision system. The framework minimizes a loss function between the target image and the retinal plane observation, employing a nonlinear conjugate gradient (NCG) method for optimization. The proposed framework's feasibility is demonstrated through numerical simulations and optical experiments.

We have achieved real-time generation up to 32K full-color rainbow holograms using a standard PC. Our previous work with 8K resolution has been extended the calculation to support resolutions up to 32K. The hologram generation utilizes a line source approximation, which 1.6 to 2.3 times faster than the previous slit model used for image hologram calculation. Experimental results with 140,000 object points demonstrate that holograms can be generated and displayed at approximately 18 frames per second for 8K resolution, 10 frames per second for 16K, and 2.6 frames per second for 32K resolution.

Waveguide holography has been actively researched for augmented reality (AR) near-eye displays (NED). However, observing computer-generated hologram (CGH) images representing three-dimensional (3D) scenes in the waveguide platform reveals challenges such as multiple 3D image replications and a very limited eye-box. In this presentation, we propose the anamorphic computer-generated hologram (CGH) synthesis algorithm that leverages the structural characteristics of waveguide type NED to simultaneously expand the eye-box and allow the user to perceive the depth of the image without multiple image replication. Furthermore, using the wave propagation method (WPM), we analyze the phase distribution over a large area for a real-sized waveguide-based near-eye display structure and provide precise numerical simulations that agree to experimental observation of anamorphic CGHs.

We report on an effort reproduce and extend prior work on light sheets for continuous-depth holography, our Electro-holography research group at BYU has endeavored to understand and replicate their work. We reproduce their results and explore phase retrieval alternatives in an effort to incorporate the precise, three dimensional control of light. We investigate sensitive parameters both in the physical optics and the code to generate high contrast, high power holographic light sheets.

The proposed wavelength multiplexed volume holographic optical coupler uses the total reflection phenomenon to guide the incident light in Bayfol HX 200TM photopolymer to improve solar collection. Multiple wavelength angular and spectral characterizations are presented. The novelty of the device is the particular combination of diffractive optical elements designed so that they can be holographically recorded at 457nm and 532nm without using prisms and will function to couple three primary colours (Blue, Green, and Red) of the visible spectrum in a single device.

We present a method to fabricate volume holograms with enhanced efficiency using (3+1)D printing. This technique incorporates dynamic light exposure with traditional 3D laser writing, allowing precise control of each voxel's refractive index. This innovation overcomes the traditional 1/M² efficiency limit of volume holograms, achieving a 1/M relationship by digitally filtering undesired interference terms. Experimentally, we demonstrated this linear diffraction efficiency relation by fabricating up to M=50 volume gratings. This advancement has implications for optical storage, interconnects, and other photonic applications, paving the way for more efficient holographic systems.

Bayfol® HX photopolymer films prove themselves as easy-to-process, full color recording materials for volume holographic optical elements (vHOEs) and are available at industrial scale. Bayfol® HX is compatible to plastic processing techniques like thermoforming, film insert molding, and casting. See through applications such as, HMD and HUD, have demanding performance requirements on combiner and imaging technologies such as efficiency, optical function, and clarity. The properties of Bayfol® HX make it principally well suited to solve these challenges in primary display, and near-infrared imaging applications such as eye-tracking. Based on a customizable and available toolbox, Bayfol® HX can be adopted for a variety of such applications. To serve further sensing applications we introduced our standardized Bayfol® HX203 film sensitized in the NIR - comparable to Bayfol® HX200 sensitized for RGB. Besides previously reported demonstration examples with this new NIR sensitized Bayfol® HX film we report on a further eye-tracking architecture which can be realized.

This study explored the optimization of multi-layered diffractive optical elements (DOEs) for compressive sensing in imaging systems. We investigated how increasing the number of DOE layers affected the coherence value of the system's transmission matrix, a key performance indicator for image reconstruction in compressive sensing. Through wave-optics simulations, we demonstrated that coherence decreased monotonically up to 8 layers before increasing, suggesting an optimal layer count exists. We also showed that binary-structured DOEs, which are more practical to implement, could achieve near-continuous performance with proper optimization. Our findings indicated that optimized multi-layer DOEs can significantly enhance the performance of compressive sensing applications, potentially improving accuracy in snapshot super-resolution and multidimensional imaging.

Diffractive neural networks (DNNs) have proven to be a valuable tool for laser beam shaping. By treating optical systems of cascaded phase masks as physical neural networks, DNNs enable many advantageous functionalities like shaping of extended beam volumes or the simultaneous optimization of amplitude and phase. While conventional training techniques provide excellent results for an accurately assembled system, they can become unreliable when introducing misalignments into the experimental setup, leading to a longer installation time and increased susceptibility to perturbations. Here, we discuss how the sensitivity to misalignments is drastically reduced by choosing mathematical adaptations and regularization techniques motivated from both physical considerations and established machine learning methods. We show experimentally how this can be realized by using one or multiple cascaded spatial light modulators (SLMs) including a full correction of pixel crosstalk and direct reflections that cause deteriorating effects in many SLM beam shaping applications.

Phase-shifted volume Bragg gratings with high transmission peak and diffraction efficiency have been realized via interferometric phase-mask technique using PTR glass. This technique enables exotic configurations beyond that of a phase-shifted reflection Bragg grating. When the phase-shift is encoded into a chirped grating, a wavelength-tunable narrow-line filter element is generated. Transversally and longitudinally chirped, π-phase-shifted gratings have been fabricated and their properties and potential uses are explored.

Abstract: A volume Bragg grating (VBG) is used to spectrally lock a blue laser diode externally to achieve linewidth on the order of tens of picometers. In conventional laser diodes the emission linewidth tends to be on the order of several nanometers due to the short photon cavity lifetime, which is a limitation for many applications. In this work, we manufacture a VBG that is holographically recorded to form a notch reflecting filter and is used to externally lock a blue laser diode reducing its emission to tens of picometers. This external cavity locking method provides narrow laser linewidths from conventional laser blue diodes while keeping a compact system design. We extended the approach to coherently combine several blue diodes in order to increase the total output power.

Analysis of the transmission characteristics of multimode fibers (MMF) plays an increasing role in communications, biomedicine, metrology, endo-microscopy, fiber sensors, displays, and nondestructive testing. With prior knowledge of the spatial modes of MMF, mode decomposition (MD) is an effective technique that can analyze the light field at the output end of the fiber in the mode domain to obtain complete mode weights from the speckle patterns. Digital holography as standard approach retrieves both the amplitude and phase of coherent multimode beams. However, the necessary reference beam represents a major challenge, especially for long fibers, since phase noise in fibers increases with length. A number of reference-less methods have been proposed, including the Gerchberg-Saxton optimization. We propose the merging of physical models with AI, a pre-training-based physics-informed neural network for MD of MMF. Paradigm shifts for secure data transmission by classical and quantum communication with MMF are promising as well as advancements are possible for lensless MMF endoscopy, photonics chips using MMF links, neuromorphic computing, quantum computing, and reservoir computing.

This paper reports updates on the Hunt for the Hologram program, an educational initiative engaging students in university-led research to improve 3D displays through photophoretic trapping. Implemented in five classrooms with "UFO" Kits, students have trapped particles like graphite and yeast, bridging theoretical knowledge and practical application. The program, involving hands-on learning, significantly advanced students' understanding of optical trapping and contributed valuable data for 3D display technologies, enhancing both education and research in foundational 3D display technologies.

HoloLab is a holography laboratory located at the Fábrica Science Center, in Aveiro. This Science Center is managed by the University of Aveiro and belongs to the National Network of Ciência Viva Science Centers in Portugal. Fábrica is the only entity in Portugal that is open every day and all year round, that has an open holography laboratory, dedicated to public engagement in holography and laser optics. Each year the educational program, holography4all, is developed to involve participants in a variety of holography activities. This program is focused on scientific content, technological applications and artistic creation. Holography4all is organized into four types of initiatives: workshops, exhibitions, summer bootcamps and holo events. This paper will introduce the detailed holography program. Impact results, public engagement and hologram production will be presented and analyzed. Some conclusions and remarks regarding future work on the holography4all program will be discussed.

Semiconductor materials are of high scientific interest for the realization of laser and amplifier structures. With advancements in the theoretical and practical understanding of those structures, more efficient and faster devices can be developed. The gain and the linewidth enhancement factor (LEF) are decisive parameters for the performance of the device. Among other things, they influence the filamentation, chirping, and laser dynamics of the device. In this work, we present a novel methodology for the measurement of gain and LEF of laser structures. This methodology is based on holography which enables spatial and spectral resolution of the gain parameters. Therefore, this method differs from conventional methods, which only allow spectral resolution. For this reason, conventional methods usually study laser structures emitting in single lateral modes due to their missing spatial resolution. The spatial resolution is a very promising candidate for laser structures with complex lateral structures which emit in more than one lateral mode. Therefore, this method provides new insight for the practical characterization of laser devices.

This study aims to construct a dense top-hat focusing array by convolving the top-hat region into a multi-spot pattern. In simulations to array the top-hat in tiles, we verify its effectiveness by the summation of tiling holograms and top-hat generating holograms. Additionally, we discuss how the relative angle between the rotation angle of the top-hat shape and the unit cell in the square lattice affects the diffraction light intensities.

Recently, research on computer-generated hologram (CGH) has attracted attention. There is a hidden surface removal method based on a point-based method, which is a type of rendering technique in CGH, that has been extended to mirror image representation on planes or Bézier curved surfaces, but the method for Bézier curved surfaces is based on Bézier clipping and the algorithm is complicated. In this study, the calculation is simplified by dividing the curved surface into polygons and placing the point light source at symmetric positions. In addition, the calculation time is reduced by using the NVIDIA OptiX ray tracing engine for the hidden surface removal by ray tracing. This method significantly reduces calculation time, which takes only a few seconds to calculate 10,000 points, compared to the traditional method that takes several hours to calculate 6,000 points. Furthermore, we aim to achieve full-color representation for more realistic depictions.

In recent years, with the development of stereoscopic display technologies such as VR, it has become possible to experience and display as if you were in a different place. These technologies have disadvantages such as getting drunk and tired from moving after a long period of experience. Electrical holography based on computer-generated holograms (CGHs) is being studied as an ideal stereoscopic image display technology that does not get tired even after watching images for a long time. Previous studies have used this technology and head-mounted displays to create holograms based on real data using RGBD cameras and display them on head-mounted displays. In addition to this, a motion sensor is mounted on the head mount display to make the video correspond to the viewpoint movement of the observer.

Phase-only computer-generated holography is calculated by the inverse Fourier transform, rendering a fully complex plane of the hologram. Spatial light modulators used to display holograms operate in the real domain so reconstruction with minimized quantization errors inherently leads to a less optimum replay field. This study explores the use of neural networks as a proof of concept for computer generated holography by first determining if pattern recognition in the complex hologram plane is possible. Then, we use neural networks to classify holograms based on pattern recognition in the hologram plane to their respective replay fields. We show that neural networks are capable of operating in the complex hologram plane by recognizing patterns and classifying holograms based on recognized patterns to their respective replay field. The results are verified experimentally using an optical setup and using the NIST digit database for training, testing, and validating. The advance of machine learning in holography has promise of adding another degree of freedom to computer-generated holography in conjunction with conventional algorithms (e.g. Gerchberg-Saxton).

The wavefront printer is an equipment for recording wavefront generated by spatial light modulators (SLM). This printer can directly fabricate a computer-generated volume hologram (CGVH) and a holographic optical element without any transfer step. As a 3D printer, CGVHs printed by the wavefront printer reconstruct the 3D images without vergence-accommodation conflicts unlike conventional holographic printers that reconstruct holographic stereograms. A Denisyuk-type wavefront printer we have proposed records wavefronts generated by the SLM that is irradiated by light transmitted through the recording material. As a result, the optical system is almost lossless in the light intensity. However, it is very difficult to adjust the unique optical system. This paper proposes a new computational technique for correcting aberration caused by alignment errors in the optical system. We demonstrate large-scale CGVHs printed using the proposed wavefront printer, which reconstructs synthetic wavefront with tens of billion sample points.

In this paper, we present an experimental analysis of the holographically controlled translational motion of optically trapped microscopic beads using a high-speed ferroelectric liquid crystal spatial light modulator (SLM). The SLM displays binary holograms at a refresh rate of about 1.5 KHz and facilitates highly accurate movement of the beam as well as the trapped beam at a temporal interval of 100s of microseconds, offering an in-depth examination of the accuracy and constraints associated with the linear bead movement. The results highlight the advantages and limitation of bead movement using a high speed holographic display.