arXiv subjects

Mooseok Jang

Publications and source records attributed to Mooseok Jang.

11 recordsLinked to original sources

arXiv2026

Random-mapped intensity optical neural network: all-optical two-layer computing for multimodal optical-field inference

Free-space optical neural networks offer distinct advantages for computational imaging and machine vision because they can compute directly on incident optical fields. However, conventional ONNs composed of cascaded linear optical components are bound to a general linear input-output relation with square-law detection between the input field $\mathbf{x}$ and output score $\mathbf{y}$, $\mathbf{y}=|\mathbf{T}\mathbf{x}|^2$, where at best every complex-valued element of the transmission matrix $\mathbf{T}$ is trainable. This restricts each output score to a quadratic form $\mathbf{x}^\dagger\mathbf{A}\mathbf{x}$ with rank-one decision matrix $\mathbf{A}=\mathbf{t}^\dagger\mathbf{t}$. Here, we present a random-mapped intensity optical neural network (RMI-ONN) as an all-optical two-layer computational network that lifts this rank-one limit. We numerically demonstrate that a high-dimensional feature projection by a disordered medium, a programmable nonnegative intensity mask, and segmented spatial power summation together can surpass the rank-one ceiling through the expressivity of higher-rank quadratic decision boundaries. Furthermore, exploiting the vectorial coherent wave-mixing nature of the disordered medium, we experimentally validate multimodal classification of amplitude, phase, and polarization on MNIST, Fashion-MNIST, and Quick Draw with the RMI-ONN, under a single optical configuration across all encoding domains. These results provide a practical and conceptual basis for scalable direct-field optical processors capable of exploiting amplitude, phase, and polarization information within a unified intensity-based inference framework.

physics.optics↗

arXiv2026

Feature-domain Fourier ptychographic tomography with dark-field illumination

Fourier ptychographic tomography (FPT) is an implementation of intensity diffraction tomography that reconstructs three-dimensional (3D) refractive-index (RI) distributions from angle-varied intensity measurements. The distinctive advantage of FPT emerges when incorporating dark-field illumination, which extends the space-bandwidth product toward gigavoxel-scale volumetric imaging, yet dark-field measurements are highly sensitive to system imperfections and often have low signal-to-noise ratios. Here, we propose feature-domain FPT (FD-FPT), which evaluates data fidelity after feature extraction and is optimized using automatic differentiation. In numerical and experimental tests, FD-FPT resolves structures near the synthetic-aperture cutoff far more reliably than the spatial-domain baseline (SD-FPT). Notably, in a whole-mount Oedogonium specimen, the reticulate chloroplast network and transverse septa were resolved only by FD-FPT. We further demonstrate a 1.81-gigavoxel RI reconstruction of a mouse adrenal gland section across a 1.66 by 1.40 square millimeter field of view, establishing FD-FPT as a practical route to label-free volumetric imaging that combines millimeter-scale coverage with cellular-scale structural contrast.

physics.optics↗

arXiv2026

On the limits and opportunities of AI reviewers: Reviewing the reviews of Nature-family papers with 45 expert scientists

With the advancement of AI capabilities, AI reviewers are beginning to be deployed in scientific peer review, yet their capability and credibility remain in question: many scientists simply view them as probabilistic systems without the expertise to evaluate research, while other researchers are more optimistic about their readiness without concrete evidence. Understanding what AI reviewers do well, where they fall short, and what challenges remain is essential. However, existing evaluations of AI reviewers have focused on whether their verdicts match human verdicts (e.g., score alignment, acceptance prediction), which is insufficient to characterize their capabilities and limits. In this paper, we close this gap through a large-scale expert annotation study, in which 45 domain scientists in Physical, Biological, and Health Sciences spent 469 hours rating 2,960 individual criticisms (each targeting one specific aspect of a paper) from human-written and AI-generated reviews of 82 Nature-family papers on correctness, significance, and sufficiency of evidence. On a composite of all three dimensions, a reviewing agent powered by GPT-5.2 scores above each paper's top-rated human reviewer (60.0% vs. 48.2%, p = 0.009), while all three AI reviewers (including Gemini 3.0 Pro and Claude Opus 4.5) exceed the lowest-rated human across every dimension. AI reviewers' accurate criticisms are also more often rated significant and well-evidenced, and surface a distinct 26% of issues no human raises. However, AI reviewers overlap far more than humans do (21% vs. 3% for cross-reviewer pairs), and exhibit 16 recurring weaknesses humans do not share, such as limited subfield knowledge, lack of long context management over multiple files, and overly critical stance on minor issues. Overall, our results position current AI reviewers as complements to, not substitutes for, human reviewers.

cs.CL↗

arXiv2025

Generalizable Holographic Reconstruction via Amplitude-Only Diffusion Priors

Phase retrieval in inline holography is a fundamental yet ill-posed inverse problem due to the nonlinear coupling between amplitude and phase in coherent imaging. We present a novel off-the-shelf solution that leverages a diffusion model trained solely on object amplitude to recover both amplitude and phase from diffraction intensities. Using a predictor-corrector sampling framework with separate likelihood gradients for amplitude and phase, our method enables complex field reconstruction without requiring ground-truth phase data for training. We validate the proposed approach through extensive simulations and experiments, demonstrating robust generalization across diverse object shapes, imaging system configurations, and modalities, including lensless setups. Notably, a diffusion prior trained on simple amplitude data (e.g., polystyrene beads) successfully reconstructs complex biological tissue structures, highlighting the method's adaptability. This framework provides a cost-effective, generalizable solution for nonlinear inverse problems in computational imaging, and establishes a foundation for broader coherent imaging applications beyond holography.

physics.optics↗

arXiv2025

Physics-Aware Style Transfer for Adaptive Holographic Reconstruction

Inline holographic imaging presents an ill-posed inverse problem of reconstructing objects' complex amplitude from recorded diffraction patterns. Although recent deep learning approaches have shown promise over classical phase retrieval algorithms, they often require high-quality ground truth datasets of complex amplitude maps to achieve a statistical inverse mapping operation between the two domains. Here, we present a physics-aware style transfer approach that interprets the object-to-sensor distance as an implicit style within diffraction patterns. Using the style domain as the intermediate domain to construct cyclic image translation, we show that the inverse mapping operation can be learned in an adaptive manner only with datasets composed of intensity measurements. We further demonstrate its biomedical applicability by reconstructing the morphology of dynamically flowing red blood cells, highlighting its potential for real-time, label-free imaging. As a framework that leverages physical cues inherently embedded in measurements, the presented method offers a practical learning strategy for imaging applications where ground truth is difficult or impossible to obtain.

physics.optics↗

arXiv2023

Large-volume focus control at 10 MHz refresh rate via fast line-scanning amplitude-encoded scattering-assisted holography

The capability of focus control has been central to optical technologies that require both high temporal and spatial resolutions. However, existing varifocal lens schemes are commonly limited to the response time on the microsecond timescale and share the fundamental trade-off between the response time and the tuning power. Here, we propose an ultrafast holographic focusing method enabled by translating the speed of a fast 1D beam scanner into the speed of the complex wavefront modulation of a relatively slow 2D spatial light modulator. Using a pair of a digital micromirror device and a resonant scanner, we demonstrate an unprecedented refresh rate of focus control of 31 MHz, which is more than 1,000 times faster than the switching rate of a digital micromirror device. We also show that multiple micrometer sized focal spots can be independently addressed in a range of over 1 MHz within a large volume of 5 mm x 5 mm x 5.5 mm, validating the superior spatiotemporal characteristics of the proposed technique - high temporal and spatial precision, high tuning power, and random accessibility in a three-dimensional space. The demonstrated scheme offers a new route towards three-dimensional light manipulation in the 100 MHz regime.

physics.optics↗

arXiv2022

Acousto-optic volumetric gating for reflection-mode deep optical imaging within a scattering medium

The imaging depth of deep-tissue optical microscopy is governed by the performance of the gating operation that suppresses the multiply scattered waves obscuring the ballistic waves. Although various gating operations based on confocal, time-resolved/coherence-gated, and polarization-selective detections have proven to be effective, each has its own limitation; certain types of multiply scattered waves can bypass the gating. Here, we propose a method, volumetric gating, that introduces ultrasound focus to confocal reflectance imaging to suppress the multiply scattered waves traveling outside the ultrasonic focal volume. The volumetric gating axially rejects the multiply scattered wave traveling to a depth shallower than the object plane while suppressing the deeper penetrating portion that travels across the object plane outside the transversal extent of the ultrasonic focus of 30${\times}$90$ μm^2$. These joint gating actions along the axial and lateral directions attenuate the multiply scattered waves by a factor of 1/1000 or smaller, thereby extending the imaging depth to 12.1 times the scattering mean free path while maintaining the diffraction-limited resolution of 1.5 $μ$m. We demonstrated an increase in the imaging depth and contrast for internal tissue imaging of mouse colon and small intestine through their outer walls. We further developed theoretical and experimental frameworks to characterize the axial distribution of light trajectories inside scattering media. The volumetric gating will serve as an important addition to deep-tissue imaging modalities and a useful tool for studying wave propagation in scattering media.

physics.optics↗

arXiv2020

DeepPhaseCut: Deep Relaxation in Phase for Unsupervised Fourier Phase Retrieval

Fourier phase retrieval is a classical problem of restoring a signal only from the measured magnitude of its Fourier transform. Although Fienup-type algorithms, which use prior knowledge in both spatial and Fourier domains, have been widely used in practice, they can often stall in local minima. Modern methods such as PhaseLift and PhaseCut may offer performance guarantees with the help of convex relaxation. However, these algorithms are usually computationally intensive for practical use. To address this problem, we propose a novel, unsupervised, feed-forward neural network for Fourier phase retrieval which enables immediate high quality reconstruction. Unlike the existing deep learning approaches that use a neural network as a regularization term or an end-to-end blackbox model for supervised training, our algorithm is a feed-forward neural network implementation of PhaseCut algorithm in an unsupervised learning framework. Specifically, our network is composed of two generators: one for the phase estimation using PhaseCut loss, followed by another generator for image reconstruction, all of which are trained simultaneously using a cycleGAN framework without matched data. The link to the classical Fienup-type algorithms and the recent symmetry-breaking learning approach is also revealed. Extensive experiments demonstrate that the proposed method outperforms all existing approaches in Fourier phase retrieval problems.

cs.CV↗

arXiv2018

Coherent space-gated microscopy: a step towards deep-tissue phase imaging of biological cells

High-resolution optical microscopy suffers from a low contrast in scattering media where a multiply scattered wave obscures a ballistic wave used for image formation. To extend the imaging depth, various gating operations - confocal, coherence, and polarization gating - have been devised to filter out the multiply scattered wave. However, these gating methods are imperfect as they all act on the detection plane located outside a scattering medium. Here, we present a new gating scheme, called 'space' gating, that rejects the multiply scattered wave directly at the object plane inside a scattering medium. Specifically, we introduced a 30 $μ$m-wide acoustic focus to the object plane and reconstructed a coherent image only with the ballistic wave modulated by acousto-optic interaction. This method allows us to reject the multiply scattered wave that the existing gating methods cannot filter out and improves the ratio of the ballistic wave to the multiply scattered wave by more than 100 times for a scattering medium more than 20 times thicker than its scattering mean free path. Using the coherent imaging technique based on space gating, we demonstrate the unprecedented imaging capability - phase imaging of optically transparent biological cells fully embedded within a scattering medium - with a spatial resolution of 1.5 $μ$m.

physics.optics↗

arXiv2017

Complex wavefront engineering with disorder-engineered metasurfaces

Recently, complex wavefront engineering with disordered media has demonstrated optical manipulation capabilities beyond those of conventional optics. These capabilities include extended volume, aberration-free focusing and subwavelength focusing via evanescent mode coupling. However, translating these capabilities to useful applications has remained challenging as the input-output characteristics of the disordered media ($P$ variables) need to be exhaustively determined via $O(P)$ measurements. Here, we propose a paradigm shift where the disorder is specifically designed so that its exact characteristics are known, resulting in an a priori determined transmission matrix that can be utilized with only a few alignment steps. We implement this concept with a disorder-engineered metasurface, which exhibits additional unique features for complex wavefront engineering such as an unprecedented optical memory effect range, excellent stability, and a tailorable angular scattering profile.

physics.optics↗

arXiv2015

Optical focusing inside scattering media with time-reversed ultrasound microbubble encoded (TRUME) light

Focusing light inside scattering media in a freely addressable fashion is challenging, as the wavefront of the scattered light is highly disordered. Recently developed ultrasound-guided wavefront shaping methods are addressing this challenge, albeit with relatively low modulation efficiency and resolution limitations. In this paper, we present a new technique, time-reversed ultrasound microbubble encoded (TRUME) optical focusing, which is able to focus light with improved efficiency and sub-ultrasound wavelength resolution. This method ultrasonically destructs microbubbles, and measures the wavefront change to compute and render a suitable time-reversed wavefront solution for focusing. We demonstrate that the TRUME technique can create an optical focus at the site of bubble destruction with a size of ~2 microns. Due to the nonlinear pressure-to-destruction response, the TRUME technique can break the addressable focus resolution barrier imposed by the ultrasound focus. We experimentally demonstrate a 2-fold addressable focus resolution improvement in a microbubble aggregate target.

physics.optics↗