XJXin Jin

Xin Jin

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xin Jin has a total of 145 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    LFCD-Net: physics-inspired 3D reconstruction architecture for miniature light-field microscopy

    by JingFei Hou, Yue Xing, Chuqi Yuan, Puxiang Lai, Xin Jin

    Miniature light-field microscopy is a vital tool for high-speed volumetric imaging in freely moving animals, yet its spatial resolution is constrained by the insufficient sampling inherent in simultaneous spatial-angular acquisition. To address this, we propose a physics inspired light-field characteristic driven 3D reconstruction network integrating three core innovations: spatial-angular feature blocks for aliasing suppression, multi-scale feature blocks for structural fidelity, and a physics-inspired adaptive weighting loss to ensure high-quality reconstruction of sparse biological signals. These modules tackle the sampling deficiency to effectively restore spatial resolution. Crucially, we developed the first miniature dual-path imaging system to provide high-quality paired training data. Simulations demonstrate that our method achieves a lateral resolution of 5.69   μ m , representing a 22% improvement over state-of-the-art deep learning methods and significantly outperforming traditional physics-based algorithms. Validations on complex biological structures and physical experiments across a large 500   μ m depth-of-field confirm superior axial stability and reconstruction accuracy, providing a reliable technical pathway for high-resolution in vivo 3D imaging.

  • Multi-view focal stack supervised hologram optimization based on cross constraints

    by Junjie Wei, Hongkun Cao, Xin Jin

    Holographic near-eye display (HNED) is currently a promising technology capable of providing the human eye with all the three-dimensional (3D) visual cues required for viewing. However, existing computer-generated hologram (CGH) methods have not yet achieved satisfactory visual comfort in HNED. Specifically, eyes with different states (size, position, and depth) fail to perceive an immersive 3D experience within the eyebox. This paper proposes a hologram optimization method supervised by a multi-view focal stack (MVFS) based on cross constraints. By optimizing the phase of the hologram jointly considering the low speckle, accurate focus-defocus effect, and view-consistent energy distribution, the free 3D motion of eyes with random size can be allowed during observation with perceived high-fidelity perspective images. The effectiveness of the proposed method is verified through numerical simulations and optical experiments.

  • Open Access

    From performance to structure: a comprehensive survey of advanced metasurface design for next-generation imaging

    by Yunhui Zeng, Haopeng Zhong, Zhenwei Long, Hongkun Cao, Xin Jin


  • Open Access

    Customized Scattering‐Robust 3D Speckle Correlation for Improved Resolution and Imaging Depth in Scattering Tissue

    by Yue Xing, Yuqi Yang, Dongyu Du, Ying Li, Xin Jin

    Seeing deep and clear in thick scattering tissues using fluorescence imaging is vital for exploring biological phenomena. Although various speckle illumination techniques have been developed to customize speckles to improve imaging resolution and depth in scattering tissues, the lack of a guiding physical model and well‐defined mechanisms for maintaining customized speckle statistics during propagation in scattering media limits their ability to guarantee quality improvement throughout the scattering volume. Here, the speckle with the generalized stationary distribution that can maintain the customized intensity correlation in 3D space is discovered through the derivation and analysis of speckle correlation in Fresnel propagation. The illumination system's correlation transmission model mapping the phase field to speckle intensity correlation is established, enabling the customization of speckles with consistent and expected intensity correlation in 3D space over a depth range of several thousand micrometers. With the enhanced anti‐scattering fluorescence imaging ability of the customized speckle, a more than twofold improvement in 3D average resolution and a twofold improvement in imaging depth have been achieved compared to random speckle under strong optical scattering conditions like those in mouse brain tissue. Fluorescence imaging through millimeter‐thick scattering layers has also been achieved, demonstrating the advantages of high‐resolution volumetric scattering imaging.


  • A Collaborative Error-Tolerant Synchronization Framework With its Application in Multimodal Light Field System Development

    by Mingyu Liu, Zongyang Tong, Xin Jin, Zhiting Zhang, Xiaofeng Liu, Jie Luo, Zhiheng Li

    Multimodal systems integrating cameras, LiDARs, radars, and other sensors are increasingly essential for intelligent transportation due to their superior environmental perception. However, the lack of a unified synchronization mechanism across diverse sensor modalities poses challenges for system-wide synchronization, causing data misalignment, reduced acquisition frequency, and degraded perception accuracy. This paper proposes a novel multimodal synchronization framework that enables high-frequency and low-latency data acquisition to improve perception accuracy. The framework integrates multi-frequency hardware synchronization with a temporal-spatial error-tolerant model to minimize data misalignment while maximizing effective frame rates. A frame header-based time synchronizer generates corrected multi-frequency trigger signals, achieving sub-4ms synchronization across 12 sensors and 4 modalities, improving the mean Intersection over Union (mIOU) of dynamic objects by an average of 4.97% in point cloud projection and 15.50% in image registration compared to existing methods. It enables us to build a multimodal system with the highest modality count (a 3×3 camera array, a LiDAR, a radar, and a long-wave infrared camera) and frame rate, capable of capturing novel light field data, outperforming existing multimodal setups. Evaluations on our real-world dataset show Average Precision (AP) gains of 49.35% for cyclists and 39.53% for pedestrians in object detection, along with significant improvements in image translation.


  • Optimization of a light field hologram via double-phase constrained spectrum tiling regularized by angular sampling

    by Junjie Wei, Hongkun Cao, Xin Jin

    A realistic holographic near-eye display (NED) requires uniform energy distribution within the eyebox, enabling users to have an immersive visual experience at various positions within the eyebox. However, the demand for observing low-speckle viewpoint images at different locations within the eyebox remains poorly addressed. In this study, a double-phase constrained spectrum titling encoding method is proposed for optimization of the light field hologram, which suppresses the speckle in all view images efficiently. To mitigate the uncertain effect of the perspective number of the input light field to optimization, an implicit angular sampling regulation is derived, which reversely offers the criteria for angular sampling of the holographic light field. Numerical and optical experimental results demonstrate that compared with advanced related methods, the proposed method yields reconstructed images with the least speckle noise under different pupil states, enabling users to observe high-quality corresponding viewpoint images at various positions within the eyebox.


  • Open Access

    Converting a conventional camera to a super-camera: directional atmospheric scattering modeling for passive imaging in intense real-world scattering scenarios

    by Yihui Fan, Xin Jin, Shun Zou, Haiyang Yu

    Passive imaging through intense atmospheric scattering is a critical yet formidable challenge in optical imaging, with profound implications across various applications. Conventional cameras struggle under severe scattering conditions, fundamentally limiting their effectiveness. We propose a groundbreaking directional atmospheric scattering model that revolutionizes passive imaging capabilities, converting a conventional camera to a super-camera. The model precisely characterizes directional photon propagation through scattering media, transforming this historically ill-posed problem into a well-posed solution, based on which a 4D spatial-angular scattering reconstruction method is proposed, which leverages both ballistic photons and directionally resolved scattered light, without relying on any scene-specific priors, to achieve unprecedented passive imaging performance enabling color imaging through over 12 transport mean free paths at distances up to 1.76 km. Our system recovers targets contributing as little as 0.00016% of the total detected signal, enhancing a standard camera’s signal recovery capacity by nearly 200×. To validate our approach, we introduce the first-ever real-world multiperspective scattering dataset, providing a critical benchmark for future research. We mark a paradigm shift in passive imaging, offering transformative potential for real-world applications under extreme atmospheric scattering conditions.


  • Open Access

    Anchor‐controlled generative adversarial network for high‐fidelity electromagnetic and structurally diverse metasurface design

    by Yunhui Zeng, Hongkun Cao, Xin Jin

    Metasurfaces, capable of manipulating light at subwavelength scales, hold great potential for advancing optoelectronic applications. Generative models, particularly Generative Adversarial Networks (GANs), offer a promising approach for metasurface inverse design by efficiently navigating complex design spaces and capturing underlying data patterns. However, existing generative models struggle to achieve high electromagnetic fidelity and structural diversity. These challenges arise from the lack of explicit electromagnetic constraints during training, which hinders accurate structure‐to‐electromagnetic mapping, and the absence of mechanisms to handle one‐to‐many mappings dilemma, resulting in insufficient structural diversity. To address these issues, we propose the Anchor‐controlled Generative Adversarial Network (AcGAN), a novel framework that improves both electromagnetic fidelity and structural diversity. To achieve high electromagnetic fidelity, AcGAN proposes the Spectral Overlap Coefficient (SOC) for precise spectral fidelity assessment and develops AnchorNet, which provides real‐time physics‐guided feedback on electromagnetic performance to refine the structure‐to‐electromagnetic mapping. To enhance structural diversity, AcGAN incorporates a cluster‐guided controller that refines input processing and ensures multilevel spectral integration, guiding the generation process to explore multiple configurations. Empirical analysis shows that AcGAN reduces the Mean Squared Error (MSE) by 73 % compared to current state‐of‐the‐art and significantly expands the design space to generate diverse metasurface architectures that meet precise spectral demands.


  • Spatiotemporal optimization of speckle illumination for high-speed super-resolution microscopy

    by Yue Xing, Yuqi Yang, Xin Jin

    Speckle illumination microscopy offers resolution beyond the diffraction limit but typically requires capturing hundreds of speckle frames to reconstruct a single super-resolution image, which significantly reduces imaging speed. To address this, we propose a spatiotemporal optimization method for speckle illumination that improves imaging speed. This method jointly optimizes the temporal uniformity of the speckle pattern, the spatial frequency spectrum of the phase modulation field, and the spatial contrast of the speckle pattern, enabling fast and uniform excitation of the sample’s fluorescence while maintaining the fine granular distribution and high contrast of the speckle patterns. Using only five speckle frames, our method achieves the super-resolution image, resulting in a nearly 20-fold increase in imaging speed compared to conventional speckle super-resolution methods. Simulations with resolution targets and experimental measurements of cycad leaves and potato tubers show a resolution approximately 1.7 times higher than the diffraction limit. Additionally, compared to the most advanced fast speckle illumination super-resolution methods (pseudo-random illumination and USRNN), our method exhibits superior robustness against aberration-induced illumination distortions, showcasing its robust, high-speed, and super-resolution capabilities.


  • End-to-end design of the metalens array imaging system with extended depth of field

    by Zhenwei Long, Yunhui Zeng, Xin Jin

    The introduction of the metalens array brings technological innovation and performance enhancements to light field imaging systems. While end-to-end design methods overcome the limitations imposed by the manual design of metalens array functionalities, exploration of their potential in achieving an extended depth of field remains unexplored, to the best of our knowledge. This paper proposes a framework for the end-to-end design of a metalens array imaging system tailored for extended depth of field imaging tasks, comprising an image formation model, an image reconstruction model and a specialized training algorithm. This framework effectively covers the extended depth range by leveraging depth importance sampling. The proposed image enhancement network utilizes a multi-scale residual convolution network incorporating channel aggregation and channel attention modules to obtain the final reconstruction results. The end-to-end training process is guided by a composite loss function and employs an alternating training algorithm enhanced with a stop-gradient strategy. Experiments conducted under two depth of field settings demonstrate, through both quantitative metrics and qualitative visual evaluations, that the proposed framework achieves superior imaging performance. Specifically, it demonstrates PSNR improvements of 1.42 dB and 10.03 dB, respectively, when different metalens parameterizations are applied, compared to the conventional design in the 11-times depth of field setting. Ablation studies further validate the effectiveness of the individual techniques proposed. To the best of our knowledge, this is the first work to integrate end-to-end optimization with metalens array design for extended depth of field imaging, addressing the limitations of traditional approaches and improving performance over a broad depth range.