High-speed computer-generated holography using an autoencoder-based deep neural network

Optics Letters, Год журнала: 2021, Номер 46(12), С. 2908 - 2908

Опубликована: Май 18, 2021

Learning-based computer-generated holography (CGH) provides a rapid hologram generation approach for holographic displays. Supervised training requires large-scale dataset with target images and corresponding holograms. We propose an autoencoder-based neural network (holoencoder) phase-only generation. Physical diffraction propagation was incorporated into the autoencoder’s decoding part. The holoencoder can automatically learn latent encodings of holograms in unsupervised manner. proposed able to generate high-fidelity 4K resolution 0.15 s. reconstruction results validate good generalizability holoencoder, experiments show fewer speckles reconstructed image compared existing CGH algorithms.

Язык: Английский

---

Neural 3D holography

ACM Transactions on Graphics, Год журнала: 2021, Номер 40(6), С. 1 - 12

Опубликована: Дек. 1, 2021

Holographic near-eye displays promise unprecedented capabilities for virtual and augmented reality (VR/AR) systems. The image quality achieved by current holographic displays, however, is limited the wave propagation models used to simulate physical optics. We propose a neural network-parameterized plane-to-multiplane model that closes gap between physics simulation. Our automatically trained using camera feedback it outperforms related techniques in 2D plane-to-plane settings large margin. Moreover, first naturally extend 3D settings, enabling high-quality computer-generated holography novel phase regularization strategy of complex-valued field. efficacy our approach demonstrated through extensive experimental evaluation with both VR optical see-through AR display prototypes.

Язык: Английский

---

Deep learning for digital holography: a review

Optics Express, Год журнала: 2021, Номер 29(24), С. 40572 - 40572

Опубликована: Ноя. 10, 2021

Recent years have witnessed the unprecedented progress of deep learning applications in digital holography (DH). Nevertheless, there remain huge potentials how can further improve performance and enable new functionalities for DH. Here, we survey recent developments various DH powered by algorithms. This article starts with a brief introduction to holographic imaging, then summarizes most relevant techniques DH, discussions on their benefits challenges. We present case studies covering wide range problems order highlight research achievements date. provide an outlook several promising directions widen use applications.

Язык: Английский

---

Speckle-free holography with partially coherent light sources and camera-in-the-loop calibration

Science Advances, Год журнала: 2021, Номер 7(46)

Опубликована: Ноя. 12, 2021

Computer-generated holography (CGH) holds transformative potential for a wide range of applications, including direct-view, virtual and augmented reality, automotive display systems. While research on holographic displays has recently made impressive progress, image quality eye safety are fundamentally limited by the speckle introduced coherent light sources. Here, we develop an approach to CGH using partially For this purpose, devise wave propagation model that is demonstrated in conjunction with camera-in-the-loop calibration strategy. We evaluate algorithm light-emitting diodes (LEDs) superluminescent LEDs (SLEDs) demonstrate improved characteristics resulting holograms compared lasers. SLEDs particular be promising sources because their generate sharp high-contrast two-dimensional (2D) 3D images bright, safe, almost free speckle.

Язык: Английский

---

End-to-end learning of 3D phase-only holograms for holographic display

Light Science & Applications, Год журнала: 2022, Номер 11(1)

Опубликована: Авг. 3, 2022

Computer-generated holography (CGH) provides volumetric control of coherent wavefront and is fundamental to applications such as 3D displays, lithography, neural photostimulation, optical/acoustic trapping. Recently, deep learning-based methods emerged promising computational paradigms for CGH synthesis that overcome the quality-runtime tradeoff in conventional simulation/optimization-based methods. Yet, quality predicted hologram intrinsically bounded by dataset's quality. Here we introduce a new dataset, MIT-CGH-4K-V2, uses layered depth image data-efficient input two-stage supervised+unsupervised training protocol direct high-quality phase-only holograms. The proposed system also corrects vision aberration, allowing customization end-users. We experimentally show photorealistic holographic projections discuss relevant spatial light modulator calibration procedures. Our method runs real-time on consumer GPU 5 FPS an iPhone 13 Pro, drastically enhanced performance above.

Язык: Английский

---

The state-of-the-art in computer generated holography for 3D display

Deleted Journal, Год журнала: 2022, Номер 3(3), С. 1 - 1

Опубликована: Янв. 1, 2022

Holographic displays have the promise to be ultimate 3D display technology, able account for all visual cues. Recent advances in photonics and electronics gave rise high-resolution holographic prototypes, indicating that they may become widely available near future. One major challenge driving those systems is computational: computer generated holography (CGH) consists of numerically simulating diffraction, which very computationally intensive. Our goal this paper give a broad overview state-of-the-art CGH. We make classification modern CGH algorithms, we describe different algorithmic acceleration techniques, discuss latest dedicated hardware solutions indicate how evaluate perceptual quality summarize our findings, remaining challenges projections on future

Язык: Английский

---

High-contrast, speckle-free, true 3D holography via binary CGH optimization

Scientific Reports, Год журнала: 2022, Номер 12(1)

Опубликована: Фев. 18, 2022

Holography is a promising approach to implement the three-dimensional (3D) projection beyond present two-dimensional technology. True 3D holography requires abilities of arbitrary volume with high-axial resolution and independent control all voxels. However, it has been challenging true high-reconstruction quality due speckle. Here, we propose practical solution realize speckle-free, high-contrast, by combining random-phase, temporal multiplexing, binary holography, optimization. We adopt random phase for implementation achieve maximum axial fully develop high-performance hologram optimization framework minimize quantization noise, which provides accurate high-contrast reconstructions 2D as well cases. Utilizing fast operation modulation, full-color high-framerate holographic video realized while speckle noise overcome multiplexing. Our high-quality experimentally verified projecting multiple dense images simultaneously. The proposed method can be adopted in various applications where show additional demonstration that realistic VR AR near-eye displays. realization will open new path towards next generation holography.

Язык: Английский

---

Liquid lens based holographic camera for real 3D scene hologram acquisition using end-to-end physical model-driven network

Light Science & Applications, Год журнала: 2024, Номер 13(1)

Опубликована: Фев. 29, 2024

Abstract With the development of artificial intelligence, neural network provides unique opportunities for holography, such as high fidelity and dynamic calculation. How to obtain real 3D scene generate hologram in time is an urgent problem. Here, we propose a liquid lens based holographic camera acquisition using end-to-end physical model-driven (EEPMD-Net). As core component camera, first 10 mm large aperture electrowetting-based proposed by specially fabricated solution. The design ensures that multi-layers can be obtained quickly with great imaging performance. EEPMD-Net takes information input, uses two new structures encoder decoder networks realize low-noise phase generation. By comparing intensity between reconstructed image after depth fusion target scene, composite loss function constructed optimization, high-fidelity training true realized time. achieves fast generation experiment proves has advantage low noise. used display, measurement, encryption other fields.

Язык: Английский

---

Real-time holographic camera for obtaining real 3D scene hologram

Light Science & Applications, Год журнала: 2025, Номер 14(1)

Опубликована: Фев. 8, 2025

Abstract As a frontier technology, holography has important research values in fields such as bio-micrographic imaging, light field modulation and data storage. However, the real-time acquisition of 3D scenes high-fidelity reconstruction technology not yet made breakthrough, which seriously hindered development holography. Here, novel holographic camera is proposed to solve above inherent problems completely. The consists end calculation end. At camera, specially configured liquid materials lens structure based on voice-coil motor-driving are used produce so that can quickly capture focus stack real scene within 15 ms. end, new structured network (FS-Net) designed for hologram calculation. After training FS-Net with renderer learnable Zernike phase, it enables 13 first device achieve incoherent scene, our breaks technical bottlenecks difficulty acquiring low quality reconstructed image, incorrect defocus blur. experimental results demonstrate effectiveness focal plane information scene. opens up way application display, modulation, measurement.

Язык: Английский

---

Polarization in diffractive optics and metasurfaces

Advances in Optics and Photonics, Год журнала: 2021, Номер 13(4), С. 836 - 836

Опубликована: Ноя. 2, 2021

Polarization, the path traced by light’s electric field vector, appears in all areas of optics. In recent decades, various technologies have enabled precise control polarization state, even on a subwavelength scale, at optical frequencies. this review, we provide thorough, high-level review fundamentals optics and detail how Jones calculus, alongside Fourier optics, can be used to analyze, classify, compare these elements. We work area across multiple research areas, including developments metasurfaces. This unifies large body spatially varying may interest both researchers designers systems more generally.

Язык: Английский

---