publications

2025

1. 

   Motion Hologram: Jointly optimized hologram generation and motion planning for photorealistic 3D displays via reinforcement learning

   Zhenxing Dong, Yuye Ling, Yan Li, and 1 more author

   Science Advances, 2025

   Abs DOI HTML PDF

   Holography is capable of rendering three-dimensional scenes with full-depth control and delivering transformative experiences across numerous domains, including virtual and augmented reality, education, and communication. However, traditional holography presents 3D scenes with unnatural defocus and severe speckles due to the limited space bandwidth product of the spatial light modulator (SLM). Here, we introduce Motion Hologram, a holographic technique that accurately portrays photorealistic and speckle-free 3D scenes. This technique leverages a single hologram and a learnable motion trajectory, which are jointly optimized within a deep reinforcement learning framework. Specifically, we experimentally demonstrated that the proposed technique could achieve a 4- to 5-dB PSNR improvement of focal stacks in comparison with traditional holography and could successfully depict speckle-free, high-fidelity, and full-color 3D displays using only a commercial SLM. We believe that the proposed method promises a prospective form of holographic displays that will offer immersive viewing experiences for audiences. Motion Hologram was proposed to accurately portray photorealistic and speckle-free 3D scenes via reinforcement learning.

2024

1. 

   Divide-Conquer-and-Merge: Memory- and Time-Efficient Holographic Displays

   Zhenxing Dong, Jidong Jia, Yan Li, and 1 more author

   In 2024 IEEE Conference Virtual Reality and 3D User Interfaces (VR), 2024

   DOI HTML PDF

2023

1. 

   Gaze-contingent efficient hologram compression for foveated near-eye holographic displays

   Zhenxing Dong, Yuye Ling, Chao Xu, and 2 more authors

   Displays, 2023

   Abs DOI

   Deep learning-based computer-generated holography (CGH) has developed rapidly and yielded remarkable outcomes. However, CGH rendering technologies are confronted with a significant challenge in transmitting massive holograms, which hinders the development of lightweight wearable near-eye holographic displays. Currently, hologram compression frameworks share a resemblance with image compression methods, which fail take into account the human visual system in practical near-eye displays, limiting the improvement of compression efficiency. Herein, we presented an efficient holographic compression framework based on foveated rendering, where we transmitted a high-resolution foveal region at a low compression rate and a low-resolution peripheral region at a high compression rate with dramatically reduced pixel numbers. Our method achieved a compression rate of 40× for a hologram resolution of 1024 × 1024, which represents a twofold increase in compression rate compared to the state-of-the-art (SOTA) method with a PSNR of ∼28.8 dB in the foveal image. Moreover, we further demonstrated the effectiveness of the proposed method in the optical experiment. We believe the proposed approach could be a viable remedy for the evergrowing data issue in near-eye holographic displays.

2. 

   Deep learning empowered highly compressive SS-OCT via learnable spectral–spatial sub-sampling

   Yuye Ling, Zhenxing Dong, Xueshen Li, and 2 more authors

   Optics Letters, 2023

   Abs DOI

   With the rapid advances of light source technology, the A-line imaging rate of swept-source optical coherence tomography (SS-OCT) has experienced a great increase in the past three decades. The bandwidths of data acquisition, data transfer, and data storage, which can easily reach several hundred megabytes per second, have now been considered major bottlenecks for modern SS-OCT system design. To address these issues, various compression schemes have been previously proposed. However, most of the current methods focus on enhancing the capability of the reconstruction algorithm and can only provide a data compression ratio (DCR) up to 4 without impairing the image quality. In this Letter, we proposed a novel design paradigm, in which the sub-sampling pattern for interferogram acquisition is jointly optimized with the reconstruction algorithm in an end-to-end manner. To validate the idea, we retrospectively apply the proposed method on an ex vivo human coronary optical coherence tomography (OCT) dataset. The proposed method could reach a maximum DCR of ∼62.5 with peak signal-to-noise ratio (PSNR) of 24.2 dB, while a DCR of ∼27.78 could yield a visually pleasant image with a PSNR of ∼24.6 dB. We believe the proposed system could be a viable remedy for the ever-growing data issue in SS-OCT.

3. 

   Fourier-inspired neural module for real-time and high-fidelity computer-generated holography

   Zhenxing Dong, Chao Xu, Yuye Ling, and 2 more authors

   Optics Letters, 2023

   Abs DOI

   Learning-based computer-generated holography (CGH) algorithms appear as novel alternatives to generate phase-only holograms. However, most existing learning-based approaches underperform their iterative peers regarding display quality. Here, we recognize that current convolutional neural networks have difficulty learning cross-domain tasks due to the limited receptive field. In order to overcome this limitation, we propose a Fourier-inspired neural module, which can be easily integrated into various CGH frameworks and significantly enhance the quality of reconstructed images. By explicitly leveraging Fourier transforms within the neural network architecture, the mesoscopic information within the phase-only hologram can be more handily extracted. Both simulation and experiment were performed to showcase its capability. By incorporating it into U-Net and HoloNet, the peak signal-to-noise ratio of reconstructed images is measured at 29.16 dB and 33.50 dB during the simulation, which is 4.97 dB and 1.52 dB higher than those by the baseline U-Net and HoloNet, respectively. Similar trends are observed in the experimental results. We also experimentally demonstrated that U-Net and HoloNet with the proposed module can generate a monochromatic 1080p hologram in 0.015 s and 0.020 s, respectively.

2021

1. 

   Frequency-Aware Physics-Inspired Degradation Model for Real-World Image Super-Resolution

   Zhenxing Dong, Hong Cao, Wang Shen, and 4 more authors

   arXiv, 2021

   DOI PDF