Pith. sign in

REVIEW 3 major objections 5 minor 39 references

EigenCWD: a spatially-varying deconvolution algorithm for single metalens imaging

T0 review · 3 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A spatially varying metalens blur can be inverted without building the full transfer matrix, using an eigen-decomposition of sampled point spread functions.

desk verdict A practical and reproducible space-variant deconvolution method, but the paper's central matrix decomposition is misstated and needs a serious rewrite before the scaling claims can be trusted. read the letter →

arxiv 2502.03790 v1 pith:W37GYJDP submitted 2025-02-06 physics.optics physics.comp-ph

Joel Yeo , Duane Loh , Ramon Paniagua-Dominguez , Arseniy I. Kuznetsov This is my paper · ORCID
classification physics.opticsphysics.comp-ph PACS 42.30.-d42.30.Va
keywords metalensspatiallyvaryingdeconvolutioneigenPSFdecompositionpointspreadfunctioncomputationalimagingtotalvariationregularizationADMMaberrationcorrection
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Single metalens cameras form images with aberrations that change across the field of view: a hyperbolic metalens focuses efficiently but adds coma and astigmatism off-axis, and the common Wiener filter cannot correct such spatially varying blur. This paper introduces eigenCWD, a deconvolution algorithm that treats the blur as a space-variant linear operator but never builds the full transfer matrix. Instead, it samples point spread functions across the field, decomposes them into a small set of eigen-PSFs via eigendecomposition, and interpolates the coefficients to every pixel. The resulting approximate forward model makes the inverse problem tractable, and the paper shows that eigenCWD outperforms Wiener filtering in correcting blur and geometric distortion for hyperbolic, parabolic, and spherical metalens profiles in simulation.

What carries the argument

The central object is the eigenPSF decomposition of the space-variant point spread function. From N PSFs sampled at different field positions, the paper forms a covariance matrix and takes its eigenvectors; each eigenPSF is a weighted sum of the sampled PSFs, and the eigenvector entries assigned to the sample positions are interpolated to yield field-dependent eigencoefficients. This yields the factorization H approximately equal to QA in which the forward blur is evaluated as K element-wise multiplications followed by K convolutions. The inverse problem is then solved by ADMM with total variation regularization using that factorized forward model.

What would settle it

Measure the actual point spread functions of a fabricated metalens on a fine grid across its field of view and compare them with the nominal simulation used in Eq. (15); if the real PSFs differ substantially from the model or vary non-smoothly, the reconstruction quality shown in Figs. 5 and 6 will not be reproduced on experimental data. A simpler experimental check is to image a resolution target with known background on a hyperbolic metalens and see whether eigenCWD introduces the ringing artifacts the paper attributes to finite-support violations.

Watch

Extended reading notes

Core claim

The paper claims that a spatially varying blur in metalens imaging can be efficiently inverted by factoring the transfer matrix H into a product QA, where Q holds K eigen-PSFs and A holds their interpolated coefficients, with K much smaller than the number of image pixels. The factorization, derived from an eigendecomposition of sampled point spread functions, captures the space-variant blur well enough to allow high-quality deconvolution of images from a single metalens. Using an ADMM solver with total variation regularization, eigenCWD reconstructs objects from simulated blurred images of hyperbolic, parabolic, and spherical metalenses, removing both field-dependent coma and barrel distortion that remain in Wiener-filtered results. The paper also shows that truncating the number of eigen-PSFs to about 200 for a 19 by 19 sampling grid preserves reconstruction quality while cutting computation time, demonstrating that the approach scales to realistic image sizes.

Load-bearing premise

The results assume that the real point spread functions of the metalens match the nominal phase-profile simulations and vary smoothly enough that a modest number of sampled PSFs plus interpolation can represent every pixel's blur; fabrication errors, partial coherence, or a nonzero background around the object would break this assumption, and the paper itself notes that finite-support objects are required to avoid ringing artifacts.

Editorial extensions

If this is right

  • A metalens image can be deblurred at full field of view without training data, using only a small set of measured or simulated PSFs.
  • The cost of deconvolution scales with the number of retained eigen-PSFs K rather than the number of pixels, so megapixel metalens images remain feasible.
  • Truncating the eigen-PSF components gives a controllable speed-quality tradeoff, with roughly 200 components sufficient for the hyperbolic lens studied.
  • The same algorithm corrects coma and barrel distortion for hyperbolic, parabolic, and spherical metalens profiles, so one method covers common metalens designs.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The rank K needed to represent the blur could serve as a diagnostic of space-variant complexity: lenses with severe coma require more eigen-PSFs, and a lens whose PSFs barely change would need only a handful, in which case Wiener filtering is nearly optimal.
  • The finite-support limitation points to an obvious preprocessing step: subtract the background or apply an edge-tapering mask before deconvolution, which the paper does not test but mentions as a future remedy.
  • The same eigendecomposition-of-PSFs strategy could be applied to other space-variant imaging systems, such as multimode fiber imaging or atmospheric turbulence, where the forward model is also too large to store explicitly.
  • Because the eigencoefficients are computed by interpolation, the method would likely fail at field positions where the PSF changes abruptly, e.g., near the edges of the aperture or at phase discontinuities; sampling adaptively at such positions could be tested.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes eigenCWD, a spatially-varying deconvolution algorithm for single-metalens imaging. The forward model is approximated by decomposing a set of sampled, space-variant PSFs into K eigenPSFs with interpolated eigencoefficients, yielding a forward operator expressed as a sum of K convolutions. Deconvolution is cast as an ADMM optimization with total-variation regularization. The authors simulate images from hyperbolic, parabolic, and spherical metalens phase profiles using an angular-spectrum PSF model, and demonstrate that eigenCWD removes coma and barrel distortion better than Wiener filtering. The paper includes a Zenodo code repository and explicitly acknowledges a finite-support limitation.

Significance. If the technical foundation is corrected, this is a useful contribution to computational imaging: it offers a space-variant deconvolution route that avoids the explicit M^2 by M^2 transfer matrix, with memory and computation scaling roughly as O(K M^2 log M^2) and O(K M^2) for K retained eigenPSFs. The paper's strengths are its concrete connection to the metalens aberration problem, systematic experiments over the PSF sampling density and component truncation, and the availability of runnable code. The claims are, however, demonstrated only on noiseless simulations that use the same nominal PSF model that generates the eigenPSF basis; practical superiority over Wiener filtering for real metalens images is not yet established.

major comments (3)
  1. [3, Eqs. (11)–(12)] The factorization H ≈ QA is not the operator computed by the three-step procedure described after Eq. (12). With Q in R^{M^2 x K} containing flattened eigenPSFs and A in R^{K x M^2} containing eigencoefficients, QAf equals sum_k q_k (a_k^T f), which for K=1 is a rank-one matrix q_1 a_1^T. The implemented procedure instead computes g = sum_k q_k * (a_k elementwise f), whose matrix is sum_k C_k diag(a_k), where C_k is the M^2 by M^2 convolution matrix of q_k; for K=1 this is a full-rank block-circulant matrix, not a rank-one outer product. The correct statement is H ≈ sum_k C_k diag(a_k). The memory and complexity argument should be rephrased in terms of storing K convolution kernels and K coefficient maps, rather than the sizes of Q and A.
  2. [2, Eqs. (3)–(4) and Eq. (15)] The notation in Eq. (3) writes p(u,v,x,y), but Eq. (4) substitutes q_k(x-u,y-v), indicating that x,y in Eq. (3) are offset coordinates relative to the source position. This ambiguity is the source of the algebraic inconsistency in Eq. (11). Please define explicitly whether the sampled PSFs and eigenPSFs are represented as functions of offsets or of absolute image coordinates, and keep that convention consistent throughout the matrix formulation. In Eq. (15), the expression inside the angular-spectrum propagator Pf appears to be a product of the phase factor and the lens phase phi(x,y); as printed, the notation is ambiguous and should be corrected.
  3. [4 and 5] All reconstruction results are noiseless simulations generated from the same nominal PSF model used to construct the eigenPSF basis. No noise, background, or experimental data are demonstrated; the only experimental reference, [38], is mentioned in Sec. 5 as a source of ringing artifacts without showing the result. The central claim that eigenCWD 'surpasses Wiener filtering' for single-metalens imaging should therefore be qualified to the simulated, finitely supported, noiseless setting, or supported by experiments with realistic noise and background. As written, the practical relevance claim is stronger than the evidence.
minor comments (5)
  1. [2, Eq. (5)] Please clarify whether C is the Gram matrix of mean-centered PSFs. If the mean is subtracted, Eq. (8) requires an additional mean term, because the reconstruction p_j = sum_i (w_i)_j q_i holds only when the decomposition is performed without centering or when the mean is explicitly included.
  2. [4, first paragraph] The simulation parameters contain apparent typographical errors: 'L = 1.25 mis placed s = 2 maway' should read 'L = 1.25 m is placed s = 2 m away' or similar, and the stated angular field of view of 46.9 deg does not follow from L=1.25 m and s=2 m. Please check the units and the calculation.
  3. [4.1, Figs. 3–4] The explanation for the non-monotonic PSNR is that the interpolated eigencoefficients do not conserve total pixel energy. Please state explicitly whether the forward model in the reconstruction step normalizes the effective PSFs, and whether the same effect influences the PSNR improvements reported in Fig. 5.
  4. [3, Eq. (13)] The ADMM updates are deferred to the paper by Sroubek [29]. Since the operator in Eq. (13) is a sum of convolutions rather than a standard matrix, a brief derivation of the proximal steps or a pseudocode listing would substantially improve reproducibility.
  5. [5, reference [38]] Reference [38] is cited as the source of the observed ringing artifacts, but [38] describes GaP quadratic metalenses; please clarify whether eigenCWD was actually applied to data from [38] or whether [38] is cited only as a related experimental imaging demonstration.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the only self-citation is a non-load-bearing limitation note about experimental ringing.

full rationale

The derivation chain is self-contained. EigenPSF constructs basis PSFs by eigendecomposition of the covariance of sampled PSFs (Eqs. 5-8); Eq. (8) is the exact reconstruction identity p_j = sum_i (w_i)_j q_i by orthonormality of the eigenvectors, so the decomposition is a linear-algebra identity rather than a fitted quantity renamed as a prediction. Section 4.1 validates the eigenPSF approximation against the explicit integral Eq. (2) using the same physical PSF model Eq. (15), which is an internal consistency check of the approximation; Section 4.2 deconvolves an image formed by the full integral using the eigenPSF approximate forward model, which is a genuine model-mismatch test, not a target-fitting loop. No parameter is fitted to the target image and then reported as a prediction. The only self-citation, Ref. [38] (same author group), appears in Sec. 5 to disclose a finite-support limitation (ringing artifacts observed on experimental data) and is not load-bearing for the central scaling or reconstruction claims; it is weighed as a minor, candid limitation note, which is why the score is 2 rather than 0. The loose use of QAf in Eq. (11) to denote the sum-of-convolutions operator of Eq. (4) is a notation and correctness concern about the scaling argument, but not a circular dependency: the algorithm's efficiency rests on performing K FFT-based convolutions, independent of whether H is genuinely low rank.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central contribution does not introduce new physical entities. Its free parameters are algorithmic choices, sampling density, truncation level, and regularizer weights, not fitted physical constants. The main unverified load-bearing assumption is that the simulated PSF model equals the real metalens behavior; the paper uses Eq. (15) both for the eigenPSF basis construction and for the test images, which is a self-consistency check rather than an external validation.

free parameters (4)
  • PSF sampling grid size n = 3x3 to 19x19, i.e., N = 9 to 361
    The number and placement of sampled PSFs determine approximation accuracy; the paper explores several grids but provides no automatic selection rule.
  • Number of retained eigenPSF components K = about 150 to 200 of up to 361
    Truncation level is chosen by observing SSIM saturation in Figs. 3b and 5b; no principled criterion is given.
  • Data-fidelity weight mu = 10^5
    Set by hand in Eq. (13) for all runs; no sensitivity analysis is reported.
  • Total variation regularization weight alpha = 1
    Set by hand in Eq. (13); no sensitivity analysis is reported.
assumptions (5)
  • standard math The convolution theorem applies to each term (f a_i) convolved with q_i in Eq. (4).
    Used in Sec. 2 to make the forward model computationally efficient; this is standard Fourier analysis.
  • standard math Eigendecomposition of the covariance matrix C in Eq. (6) yields orthonormal eigenPSFs, and sorted eigenvalues allow optimal truncation.
    Sec. 2, Eqs. (6)-(8); this is standard principal component / Karhunen-Loeve decomposition.
  • domain assumption Incoherent image formation is linear in the PSF, Eq. (2), and PSFs are computed from the nominal phase profile via Eq. (15).
    Real metalens PSFs may include fabrication errors, partial coherence, and defocus, which are not modeled in the simulations.
  • domain assumption PSFs vary smoothly enough over the object plane that interpolating eigencoefficients from N sampled locations accurately represents all M^2 PSF columns.
    Sec. 2 and Fig. 1; this is the core approximation behind eigenPSF and is validated only against simulated PSFs.
  • domain assumption The object is finitely supported with a negligible background.
    The paper states this limitation explicitly in Sec. 5, citing ringing artifacts when eigenCWD is applied to experimental data in [38].

how reviews work

0 comments
Cite this review

Pith. "Pith review of EigenCWD: a spatially-varying deconvolution algorithm for single metalens imaging." pith.science (2026). https://pith.science/paper/W37GYJDP

@misc{pith2026250203790,
  author       = {Pith},
  title        = {Pith review of: EigenCWD: a spatially-varying deconvolution algorithm for single metalens imaging},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/W37GYJDP}},
  note         = {Machine review of arXiv:2502.03790}
}
read the original abstract

The miniaturization of optics through the use of two-dimensional metalenses has enabled novel applications in imaging. To date, single-lens imaging remains the most common configuration, in part due to the limited focusing efficiency of metalenses. This results in limitations when it comes to wavefront manipulation and, thus, unavoidable aberrations in the formed image that require computational deconvolution to deblur the image. For certain lens profiles, such as the most common hyperbolic one that results in the highest efficiencies, at large fields of view, spatially-varying aberrations such as coma or astigmatism are prominent. These aberrations cannot be corrected for by traditional deconvolution methods, such as Wiener filtering. Here, we develop a spatially-varying deconvolution algorithm based on eigenvalue column-wise decomposition (eigenCWD). EigenCWD solves a minimization problem of the error between the measured image and the estimated image of the object to be reconstructed through an approximate forward blurring model. This approximate forward model uses an eigendecomposition of the spatially-varying point spread functions for fast computation, allowing for efficient scaling to larger image sizes and blurring kernels common in metalens imaging. We demonstrate eigenCWD's ability to correct spatially-varying blur and distortions for various lens profiles, surpassing that of the Wiener filter.

Figures

Figures reproduced from arXiv: 2502.03790 by the authors.

Figure 1
Figure 1. Schematic of the eigenPSF formalism for 25 PSFs sampled on a [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Schematic of the eigenCWD algorithm. The object is blurred using the eigenPSF forward model, and [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Approximation of forward blurring model with eigenPSF method. (a) The computed images using [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Pixel-wise difference between the computed blurred images using the eigenPSF method against evaluating [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: spatially-varying deconvolution via eigenCWD. The image to be deconvolved is simulated using the explicit [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Comparison of image deblurring using the Wiener filter and our proposed eigenCWD for (a) cameraman and [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: EigenPSF forward model for all sampled PSFs gridsizes. [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]
Figure 8
Figure 8. Figure 8: EigenCWD reconstruction for all sampled PSFs gridsizes. [PITH_FULL_IMAGE:figures/full_fig_p012_8.png]
Figure 9
Figure 9. Figure 9: Time taken to perform (left) eigenPSF forward model blurring and (right) eigenCWD reconstruction on a [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]
Figure 10
Figure 10. Figure 10: Reconstruction of cameraman for hyperbolic, parabolic and spherical lenses. All parameters of the lenses [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]
Figure 11
Figure 11. Figure 11: Reconstruction of USAF1951 for hyperbolic, parabolic and spherical lenses. All parameters of the lenses [PITH_FULL_IMAGE:figures/full_fig_p015_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 23 canonical work pages

  1. [38]

    Large field-of-view and multi-color imaging with GaP quadratic metalenses

    Anton V Baranikov, Egor Khaidarov, Emmanuel Lassalle, Damien Eschimese, Joel Yeo, N Duane Loh, Ramon Paniagua-Dominguez, and Arseniy I Kuznetsov. Large field-of-view and multi-color imaging with GaP quadratic metalenses. Laser Photon. Rev., November 2023

    2023

  2. [1]

    Roadmap for optical metasurfaces

    Arseniy I Kuznetsov, Mark L Brongersma, Jin Yao, Mu Ku Chen, Uriel Levy, Din Ping Tsai, Nikolay I Zheludev, Andrei Faraon, Amir Arbabi, Nanfang Yu, Debashis Chanda, Kenneth B Crozier, Alexander V Kildishev, Hao Wang, Joel K W Yang, Jason G Valentine, Patrice Genevet, Jonathan A Fan, Owen D Miller, Arka Majumdar, Johannes E Fröch, David Brady, Felix Heide,...

    2024

  3. [2]

    Optically resonant dielectric nanostructures

    Arseniy I Kuznetsov, Andrey E Miroshnichenko, Mark L Brongersma, Yuri S Kivshar, and Boris Luk’yanchuk. Optically resonant dielectric nanostructures. Science, 354(6314), November 2016. 9 Ghostbuster: a phase retrieval diffraction tomography algorithm for cryo-EM A PREPRINT

    work page 2016

  4. [3]

    Recent advances in planar optics: from plasmonic to dielectric metasurfaces

    Patrice Genevet, Federico Capasso, Francesco Aieta, Mohammadreza Khorasaninejad, and Robert Devlin. Recent advances in planar optics: from plasmonic to dielectric metasurfaces. Optica, OPTICA, 4, January 2017

    2017

  5. [4]

    Dielectric metalens for miniaturized imaging systems: progress and challenges

    Meiyan Pan, Yifei Fu, Mengjie Zheng, Hao Chen, Yujia Zang, Huigao Duan, Qiang Li, Min Qiu, and Yueqiang Hu. Dielectric metalens for miniaturized imaging systems: progress and challenges. Light Sci Appl, 11(1):195, June 2022

    2022

  6. [5]

    The diffractive achromat full spectrum computational imaging with diffractive optics

    Yifan Peng, Qiang Fu, Felix Heide, and Wolfgang Heidrich. The diffractive achromat full spectrum computational imaging with diffractive optics. ACM Trans. Graph., 35(4):1–11, July 2016

    work page 2016

  7. [6]

    Metasurface optics for full-color computational imaging.Science Advances, 4(2):eaar2114, February 2018

    Shane Colburn, Alan Zhan, and Arka Majumdar. Metasurface optics for full-color computational imaging.Science Advances, 4(2):eaar2114, February 2018

    work page 2018

  8. [7]

    Simultaneous achromatic and varifocal imaging with quartic metasurfaces in the visible

    Shane Colburn and Arka Majumdar. Simultaneous achromatic and varifocal imaging with quartic metasurfaces in the visible. ACS Photonics, 7(1):120–127, January 2020

    work page 2020

Show all 39 references
  1. [8]

    Design and analysis of extended depth of focus metalenses for achromatic computational imaging

    Luocheng Huang, James Whitehead, Shane Colburn, and Arka Majumdar. Design and analysis of extended depth of focus metalenses for achromatic computational imaging. Photon. Res., PRJ, 8(10):1613–1623, October 2020

    2020

  2. [9]

    Synthetic aperture metalens

    Feng Zhao, Zicheng Shen, Decheng Wang, Bijie Xu, Xiangning Chen, and Yuanmu Yang. Synthetic aperture metalens. Photonics Res., 9(12):2388, December 2021

    2021

  3. [10]

    Extrapolation, interpolation, and smoothing of stationary time series: With engineering applications

    Norbert Wiener. Extrapolation, interpolation, and smoothing of stationary time series: With engineering applications. MIT Press, London, England, June 2019

    2019

  4. [11]

    Bayesian-based iterative method of image restoration

    William Hadley Richardson. Bayesian-based iterative method of image restoration. J. Opt. Soc. Am., 62(1):55, January 1972

    1972

  5. [12]

    An iterative technique for the rectification of observed distributions

    L B Lucy. An iterative technique for the rectification of observed distributions. Astronomical Journal, V ol. 79, p. 745 (1974), 1974

    1974

  6. [13]

    Analysis of bounded variation penalty methods for ill-posed problems.Inverse Probl., 10(6):1217, December 1994

    R Acar and C R V ogel. Analysis of bounded variation penalty methods for ill-posed problems.Inverse Probl., 10(6):1217, December 1994

    1994

  7. [14]

    Learned rotationally symmetric diffractive achromat for full-spectrum computational imaging

    Xiong Dun, Hayato Ikoma, Gordon Wetzstein, Zhanshan Wang, Xinbin Cheng, and Yifan Peng. Learned rotationally symmetric diffractive achromat for full-spectrum computational imaging. Optica, OPTICA, 7(8):913– 922, August 2020

    2020

  8. [15]

    Neural nano-optics for high-quality thin lens imaging

    Ethan Tseng, Shane Colburn, James Whitehead, Luocheng Huang, Seung-Hwan Baek, Arka Majumdar, and Felix Heide. Neural nano-optics for high-quality thin lens imaging. Nat. Commun., 12(1):6493, November 2021

    2021

  9. [16]

    3D imaging using extreme dispersion in optical metasurfaces

    Shiyu Tan, Frank Yang, Vivek Boominathan, Ashok Veeraraghavan, and Gururaj V Naik. 3D imaging using extreme dispersion in optical metasurfaces. ACS Photonics, 8(5):1421–1429, May 2021

    2021

  10. [17]

    Trilobite-inspired neural nanophotonic light-field camera with extreme depth-of-field

    Qingbin Fan, Weizhu Xu, Xuemei Hu, Wenqi Zhu, Tao Yue, Cheng Zhang, Feng Yan, Lu Chen, Henri J Lezec, Yanqing Lu, Amit Agrawal, and Ting Xu. Trilobite-inspired neural nanophotonic light-field camera with extreme depth-of-field. Nat. Commun., 13(1):2130, April 2022

    2022

  11. [18]

    Deep learning enhanced achromatic imaging with a singlet flat lens

    Shanshan Hu, Xingjian Xiao, Xin Ye, Rongtao Yu, Yanhao Chu, Ji Chen, Shining Zhu, and Tao Li. Deep learning enhanced achromatic imaging with a singlet flat lens. Opt. Express, 31(21):33873–33882, October 2023

    2023

  12. [19]

    Miniature color camera via flat hybrid meta-optics

    Samuel Pinilla, Johannes E Fröch, Seyyed Reza Miri Rostami, Vladimir Katkovnik, Igor Shevkunov, Arka Majumdar, and Karen Egiazarian. Miniature color camera via flat hybrid meta-optics. Sci Adv, 9(21):eadg7297, May 2023

    2023

  13. [20]

    Achromatic imaging systems with flat lenses enabled by deep learning

    Roy Maman, Eitan Mualem, Noa Mazurski, Jacob Engelberg, and Uriel Levy. Achromatic imaging systems with flat lenses enabled by deep learning. ACS Photonics, 10(12):4494–4500, December 2023

    2023

  14. [21]

    Ultra-wide FOV meta-camera with transformer-neural-network color imaging methodology

    Yan Liu, Wen-Dong Li, Kun-Yuan Xin, Ze-Ming Chen, Zun-Yi Chen, Rui Chen, Xiao-Dong Chen, Fu-Li Zhao, Wei-Shi Zheng, and Jian-Wen Dong. Ultra-wide FOV meta-camera with transformer-neural-network color imaging methodology. AP, 6(5):056001, May 2024

    2024

  15. [22]

    Broadband achromatic imaging of a metalens with optoelectronic computing fusion

    Wei Cheng, Yan Wang, Yuqing Zhang, Huan Chen, Zhechun Lu, Fen Zhao, Yaohua Wang, Jiagui Wu, and Junbo Yang. Broadband achromatic imaging of a metalens with optoelectronic computing fusion. Nano Lett., 24(1):254–260, January 2024

    2024

  16. [23]

    Deep-learning enhanced high-quality imaging in metalens-integrated camera

    Yanxiang Zhang, Yue Wu, Chunyu Huang, Zi-Wen Zhou, Muyang Li, Zaichen Zhang, and Ji Chen. Deep-learning enhanced high-quality imaging in metalens-integrated camera. Opt. Lett., 49(10):2853–2856, May 2024

    2024

  17. [24]

    Fast iterative image restoration with a spatially varying PSF

    James G Nagy. Fast iterative image restoration with a spatially varying PSF. In Advanced Signal Processing: Algorithms, Architectures, and Implementations VII, volume 3162, pages 388–399. SPIE, October 1997. 10 Ghostbuster: a phase retrieval diffraction tomography algorithm fo...

    1997

  18. [25]

    Efficient filter flow for space-variant multiframe blind deconvolution

    Michael Hirsch, Suvrit Sra, Bernhard Schölkopf, and Stefan Harmeling. Efficient filter flow for space-variant multiframe blind deconvolution. In 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pages 607–614, June 2010

    2010

  19. [26]

    Fast approximations of Shift-Variant blur

    Loïc Denis, Eric Thiébaut, Ferréol Soulez, Jean-Marie Becker, and Rahul Mourya. Fast approximations of Shift-Variant blur. Int. J. Comput. Vis., 115(3):253–278, December 2015

    2015

  20. [27]

    A computational method for the restoration of images with an unknown, spatially-varying blur

    Johnathan Bardsley, Stuart Jefferies, James Nagy, and Robert Plemmons. A computational method for the restoration of images with an unknown, spatially-varying blur. Opt. Express, 14(5):1767–1782, March 2006

    2006

  21. [28]

    Deconvolution with a spatially variant PSF

    Tod Lauer. Deconvolution with a spatially variant PSF. In Astronomical Data Analysis II, volume 4847, pages 167–173. SPIE, December 2002

    2002

  22. [29]

    Decomposition of Space-Variant blur in image deconvolution

    Filip Sroubek, Jan Kamenicky, and Yue M Lu. Decomposition of Space-Variant blur in image deconvolution. IEEE Signal Process. Lett., 23(3):346–350, March 2016

    2016

  23. [30]

    Deconvolution for multimode fiber imaging: modeling of spatially variant PSF

    Raphaël Turcotte, Eusebiu Sutu, Carla C Schmidt, Nigel J Emptage, and Martin J Booth. Deconvolution for multimode fiber imaging: modeling of spatially variant PSF. Biomed. Opt. Express, 11(8):4759–4771, August 2020

    2020

  24. [31]

    Imaging through deconvolution with a spatially variant point spread function

    Kyle Novak and Abbie T Watnik. Imaging through deconvolution with a spatially variant point spread function. In Computational Imaging VI, volume 11731, pages 24–40. SPIE, April 2021

    2021

  25. [32]

    Distributed optimization and statistical learning via the alternating direction method of multipliers

    Stephen Boyd, Neal Parikh, Eric Chu, Borja Peleato, and Jonathan Eckstein. Distributed optimization and statistical learning via the alternating direction method of multipliers. F oundations and Trends in Machine Learning, 3(1):1–122, 2011

    2011

  26. [33]

    Anisoplanatic imaging through turbulence using principal component analysis

    Roberto Baena-Gallé, Szymon Gladysz, and Javier Mateos. Anisoplanatic imaging through turbulence using principal component analysis. In Proc. of the Advanced Maui Opt. and Space Surv. Tech. Conf.,(The Maui Economic Developt Board, Maui, 2015) , 2015

    2015

  27. [34]

    PCA-based approach for inhomogeneous PSF estimation and partial volume correction in PET

    Zacharie Irace, Anthonin Reilhac, Bruno Mendez de Vigo, Hadj Batatia, and Nicolas Costes. PCA-based approach for inhomogeneous PSF estimation and partial volume correction in PET. In 2016 IEEE Nuclear Science Symposium, Medical Imaging Conference and Room-Temperature Semicondu...

    2016

  28. [35]

    Nonlinear total variation based noise removal algorithms

    Leonid I Rudin, Stanley Osher, and Emad Fatemi. Nonlinear total variation based noise removal algorithms. Physica D, 60(1):259–268, November 1992

    1992

  29. [36]

    Band-limited angular spectrum numerical propagation method with selective scaling of observation window size and sample number

    Xiao Yu, Tang Xiahui, Qin Yingxiong, Peng Hao, and Wang Wei. Band-limited angular spectrum numerical propagation method with selective scaling of observation window size and sample number. J. Opt. Soc. Am. , 29(11):2415, November 2012

    2012

  30. [37]

    Image quality assessment: from error visibility to structural similarity

    Zhou Wang, Alan Conrad Bovik, Hamid Rahim Sheikh, and Eero P Simoncelli. Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Process., 13(4):600–612, April 2004

    2004

  31. [39]

    Reducing boundary artifacts in image deconvolution

    Renting Liu and Jiaya Jia. Reducing boundary artifacts in image deconvolution. In 2008 15th IEEE International Conference on Image Processing, pages 505–508. IEEE, 2008. 11 Ghostbuster: a phase retrieval diffraction tomography algorithm for cryo-EM A PREPRINT Supplementary Inf...

    2008

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.