Multi-Frame Blind Manifold Deconvolution for Rotating Synthetic Aperture Imaging

Dao Lin; Jian Zhang; Martin Benning

REVIEW 3 major objections 6 minor 54 references

Multi-Frame Blind Manifold Deconvolution for Rotating Synthetic Aperture Imaging

T0 review · 3 major / 6 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read Deconvolving each rotating-aperture frame, fitting a low-dimensional manifold to the results, and reconvolving before final deconvolution yields a sharper latent image than conventional multi-frame blind deconvolution.

desk verdict A coherent pipeline that combines known tools in a sensible order, but the reported gain is in-sample and the real-data case is unproven. read the letter →

arxiv 2501.19386 v1 pith:ORJGHGZA submitted 2025-01-31 stat.ME cs.CVeess.SP

Dao Lin , Jian Zhang , Martin Benning This is my paper · ORCID

classification stat.MEcs.CVeess.SP MSC 62H3594A08

keywords rotatingsyntheticapertureimagingmulti-frameblinddeconvolutionmanifoldfittinglatentimagereconstructionhyper-Laplacianpriorhalf-quadraticsplittinglow-dimensionaldeblurring

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

Rotating synthetic aperture (RSA) imaging captures several blurred views of the same scene by rotating a rectangular aperture, each view blurred by a different, unknown point-spread function. The paper tries to show that after deconvolving each frame separately, fitting a low-dimensional manifold to the deconvolved frames and locally averaging them moves those frames toward the latent sharp image, and that this denoising step, followed by reconvolution and a non-blind deconvolution, produces a sharper final reconstruction than conventional multi-frame blind deconvolution. On a simulated RGB astronaut scene with 36 blurred frames, the proposed IMR (deconvolve, manifold-fit, reconvolve) procedure raises PSNR from 25.49 dB to 28.89 dB and SSIM from 0.5778 to 0.7869 relative to the baseline. If the result holds, RSA systems could approach the resolution of larger circular apertures without their manufacturing cost, because the algorithm exploits the angle-dependent high-frequency information the rotating aperture captures.

What carries the argument

The load-bearing identity is Proposition A.1, which shows in the Fourier domain that the least-squares solution of the multi-frame convolution model is a weighted average of per-frame deconvolved images, with weights proportional to the estimated kernel power spectra. This makes the noise level of each individual deconvolved frame the bottleneck, which is exactly what the manifold-fitting step targets. The IMR procedure then runs: an IX-step (deconvolve each frame using the modified MAP blind deconvolution, with a hyper-Laplacian prior at alpha = 0.8, kernel reparameterisation by projection onto the probability simplex, and a lasso solved by Celer); an MF-step (manifold fitting in which each frame is moved toward the manifold: first estimate the contraction direction F(z_i) with distance-decaying weights, then a rank-one contraction matrix U_i, then a two-stage weighting of the u/v decomposition gives the projected point G(z_i)); and an RC-step (reconvolve the enhanced frames with estimated kernels to form denoised blurred images, then solve a final non-blind deconvolution under the same gradient prior).

What would settle it

A decisive check: run the same simulated frames but replace the manifold-fitting step with ordinary local averaging over the same r1/r2 neighbourhood with uniform weights. If the reported 3.4 dB PSNR gain over the baseline persists, the manifold geometry itself is not the active ingredient; if the gain vanishes, the contraction-direction scheme is essential.

Watch

Extended reading notes

Core claim

The central claim is that the intermediate deconvolved frames in multi-frame blind deconvolution are not just an algorithmic by-product but lie near a low-dimensional manifold that also contains the latent sharp image, so projecting those frames onto the manifold is a legitimate denoising step. The proposed IMR procedure implements this in three steps: first, a modified MAP blind deconvolution estimates per-frame blur kernels and produces deconvolved frames under a hyper-Laplacian gradient prior; second, each deconvolved frame is replaced by a weighted local average of its neighbours, with weights computed by a two-stage manifold-fitting scheme that estimates the contraction direction toward the manifold; third, the enhanced frames are reconvolved with their estimated kernels to form denoised blurred images, which are then fused by non-blind deconvolution under the same gradient prior. The paper reports that on its single simulated RSA dataset this pipeline outperforms the conventional method (PSNR 28.89 dB versus 25.49 dB; SSIM 0.7869 versus 0.5778) and also outperforms applying manifold fitting directly to the blurred frames, which the authors attribute to the deconvolved frames being more homogeneous and closer to the latent manifold.

Load-bearing premise

The load-bearing premise is that the deconvolved frames lie near a low-dimensional manifold that also contains the latent sharp image; if real RSA frames do not cluster this way, the manifold-fitting step will average away genuine detail instead of noise.

Editorial extensions

If this is right

If the manifold assumption holds on real RSA data, the same pipeline could raise the resolution of small-satellite imagery without enlarging the optics, since it exploits angle-dependent information the rotating aperture already captures.
Applying manifold fitting to deconvolved frames outperforms applying it to the raw blurred frames; the authors state this gap narrows as the number of frames grows, so the benefit is largest in the small-n regime typical of RSA capture.
The final reconstruction with the gradient prior (equation 2.17) beats the prior-free weighted Fourier average (equation 2.8), showing the hyper-Laplacian prior contributes beyond the manifold denoising step.
The framework is modular: the authors suggest deep-learning deblurring and image fusion could replace the algebraic deconvolution steps, making the manifold step a plug-in enhancement.

Reading between the lines

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

An ablation test would isolate the active ingredient: replace the two-stage manifold weights with ordinary distance-weighted local averaging over the same r1/r2 neighbourhood; if PSNR/SSIM do not drop, the contraction-direction refinement in equations (2.10)-(2.15) is not what drives the gain.
The validation is a single simulated RGB image with known kernels, and the reported PSNR fluctuates strongly when r1 is between 90 and 105; on real data, where kernel mismatch is unknown, the neighbourhood size may need re-tuning and the gains may not transfer directly.
If real RSA frames contain rotation-dependent content beyond what one shared latent manifold can represent, such as specular reflections that move with angle, the MF-step could average away genuine detail rather than noise; a targeted simulation with angle-dependent scene content would test this.
A second decomposition is possible: fix the kernels estimated by the conventional method and run only the non-blind deconvolution on the raw frames; comparing that output with the full IMR result would separate kernel-estimation gains from manifold-denoising gains.

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.

Desk Editor's Note

read the letter

Here's my take on Lin, Zhang, and Benning. The paper is a decent proof-of-concept: it takes two existing tools—Zhou et al.'s hyper-Laplacian multi-frame blind deconvolution and Yao et al.'s manifold fitting—and puts them together in a sensible order: deconvolve each frame, manifold-fit the restored frames, reconvolve with the estimated kernels, and run a final non-blind deconvolution. The ordering matters, and the paper makes a reasonable case that manifold fitting works better on deconvolved frames than on the raw blurred frames. The optimization appendices are standard and internally consistent; the derivations check out, and the Lasso reformulation for kernel estimation is clean. That is real value.

The soft spot is the evaluation, not the math. The whole empirical case is one 512x512 astronaut image, one noise realization, no real RSA data, and no code. More importantly, the headline improvement—28.89 dB vs 25.49 dB PSNR, 0.7869 vs 0.5778 SSIM—is achieved with r1 = 108, which is selected from a sweep on that same image (Figure 8 shows the optimum "near 110"). The stress-test note has this right: the reported margin is in-sample optimism, not an unbiased estimate of the method's advantage. To be fair, the paper does show a sensitivity curve over r1, which is more than many papers do, but using the tuned value on the same image for the headline comparison is a real flaw. The baseline method, Zhou et al., does not get an equivalent per-image parameter selection.

A smaller concern: the manifold hypothesis applied to deconvolved frames is plausible but untested on real RSA data. If the deconvolved frames don't cluster around the latent image—for example, because residual kernel errors create structured deviations—the averaging step could remove real detail instead of noise. The paper only indirectly acknowledges this.

Who is this for? Someone working on RSA image reconstruction or multi-frame blind deconvolution. It is a modest contribution, not a breakthrough. The math is sound and the idea is worth exploring further, but the evidence as presented is a single in-sample result. I would send it to peer review—a good referee can push for more images, real data, and honest out-of-sample tuning. I would not cite it yet for the quantitative claim.

Referee Report

3 major / 6 minor

Summary. The paper proposes a multi-frame blind manifold deconvolution procedure (IMR) for rotating synthetic aperture (RSA) imaging. The method first deconvolves each acquired frame with estimated blur kernels (IX-step), then applies a manifold-fitting denoising step to the deconvolved frames (MF-step), and finally reconvolves the enhanced frames and performs a non-blind deconvolution to estimate the latent sharp image (RC-step). The optimization is implemented via half-quadratic splitting, an iteratively reweighted least-squares scheme, and a Lasso reformulation for kernel estimation solved with the Celer solver. The numerical section evaluates the method on a single simulated RGB astronaut image corrupted by 36 synthetic kernels and Gaussian noise, reporting PSNR/SSIM gains over a conventional multi-frame blind deconvolution baseline and over direct manifold fitting on the blurred frames. The paper concludes that the proposed method can outperform conventional multi-frame blind deconvolution in pixel-intensity estimation and structural-detail preservation.

Significance. If the empirical claims hold, the paper offers a principled way to combine blind deconvolution with manifold structure, and the algorithmic machinery in Appendices A-C is a useful contribution: the frequency-domain derivation of Proposition A.1 is careful, the reformulation of kernel estimation as a positive-definite Lasso problem is non-trivial, and the three-step IX-MF-RC pipeline is clearly described. The strength of the paper is its methodological framework rather than its current evidence base, because the empirical support is limited to one image, one noise level, and no real RSA data. The reported gains are plausible but not yet established at the level claimed in Section 4.

major comments (3)

[§3.3, Fig. 8 and §3.2] The headline improvement (28.89 dB PSNR, 0.7869 SSIM, Figure 7(d)) is obtained with r1 = 108, and the same section identifies r1 = 110 as the optimal value by sweeping r1 on this single astronaut test image (Figure 8). The conventional baseline in Figure 7(a) is not given an equivalent parameter-selection procedure, and no held-out image is used to choose r1. The reported gain of +3.40 dB and +0.209 SSIM is therefore an in-sample, positively selected measure of the method's advantage, and the Section 4 claim that the method 'can outperform' conventional multi-frame blind deconvolution rests on this number. Please provide an out-of-sample evaluation, for example by selecting r1 on a separate validation image or by cross-validating over images, and report the resulting performance on held-out test images.
[§3.1–§3.3, §4] The evaluation uses a single 512×512 RGB astronaut image, one noise level (σ = 0.05), and one noise realization, with no real RSA data. The conclusion in Section 4 generalizes to RSA imaging, but the evidence is too narrow to support that generalization. Additional experiments with multiple scenes, multiple noise levels, and multiple independent noise realizations (with mean and standard deviation of PSNR/SSIM reported) would make the empirical claim credible; alternatively, the conclusions should be explicitly limited to the single simulated example.
[§2.2.2, MF-step] The load-bearing assumption of the MF-step is that the deconvolved frames {~x_i} lie near a low-dimensional manifold that also contains the latent sharp image, so that local weighted averaging (equations 2.10–2.15) moves each frame closer to the latent image. This assumption is not directly validated on real RSA data or even on a range of simulated scene contents. If the manifold structure is weak or if residual kernel errors create structured deviations rather than random noise, the MF-step could average away real detail. A concrete test would be to report, across several scenes, the change in per-frame PSNR/SSIM after manifold fitting relative to the ground truth, or to compare the method against a non-manifold denoising baseline with the same kernel estimates.

minor comments (6)

[Throughout] The word 'angel' is repeatedly used where 'angle' is intended (for example, in Section 2 and in the Introduction: 'various angels').
[Section 1, Contributions] The first contribution reads 'An muti-frame blind manifold convolution model is proposed'; it should be 'A multi-frame blind manifold deconvolution model'.
[§3.2] The text says the deconvolved images {~x_i} are obtained 'via non-blind deconvolution (solving equation (2.16))', but equation (2.16) is the RC-step reconvolution; the correct reference is equation (2.9).
[§3.3, Figure 7 discussion] The sentence 'Consequently, the reconstructed image (Figure 7 (b)) achieves superior quality compared to Figure 7 (d)' contradicts the reported metrics: Figure 7(b) has PSNR 27.02 and SSIM 0.6451, while Figure 7(d) has PSNR 28.89 and SSIM 0.7869. The comparison should be between Figure 7(b) and Figure 7(a), or the sentence should be corrected.
[Figure 2 caption] The caption says 'First row: 12 estimated blur kernels', but the experiment uses 36 kernels; please clarify that only a representative subset of 12 is displayed.
[Equations (2.8) and (2.16)] The symbol ~y_i is defined in two different places: in Section 2.1 as a denoised convolved image through F(ˆx*_i)⊙F(~k_i)=F(~y_i), and in Section 2.2.3 as ~y_i = ˆk_i ∗ ˆx*_i. These definitions are consistent only in the noiseless case; the notation should be unified or the distinction made explicit.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the IMR procedure is a genuine algorithmic composition, and the only concern (test-image hyperparameter tuning) is a statistical validity issue, not circular reasoning.

full rationale

I walked the paper's derivation chain and found no step that assumes its own conclusion. The blind deconvolution component (XK-procedure) follows the standard alternating MAP framework, and the IX-step deconvolves each frame using kernels estimated by that procedure. The MF-step applies the external manifold-fitting algorithm of Yao et al. (2023) to the deconvolved frames; this is an imported modeling assumption, not a self-citation or a disguised version of the target result. The RC-step reintroduces convolution via equation (2.16), ãƒ¢ÂÅ¡~y_i = Â¥k_i * Â¥x*_i, and then solves a fresh non-blind deconvolution problem (2.17). This is a legitimate regularized averaging/fusion of the enhanced frames: the objective uses the same estimated kernels but does not encode the ground-truth image; it only measures fidelity to the reconvolved enhanced frames plus a hyper-Laplacian prior. Proposition A.1 is a standard DFT/Parseval identity, not a circular reduction. There is no load-bearing self-citation, no imported uniqueness theorem, and no ansatz smuggled in by citation. The only substantive concern is in Section 3.3, where r1 is swept on the single astronaut test image and the reported PSNR/SSIM improvement uses a radius from that same sweep (the text says PSNR 'peaking at r1 = 110' and the final reconstruction uses r1 = 108). This is in-sample hyperparameter selection and inflates the reported gain, but it is an evaluation-bias problem, not circularity: no equation or fitted parameter is being renamed as a prediction, and the algorithm's output is not defined in terms of the evaluation metric. Under the hard rule that circularity must be exhibited by a specific reduction, the correct finding is no significant circularity.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central claim rests on a standard statistical image-formation model, the manifold hypothesis for natural images, and the convergence of half-quadratic splitting. No new physical entities are introduced. The most consequential free parameters are the regularization weights and the manifold radius r1, the latter being tuned on the evaluation image.

free parameters (5)

lambda_1 (regularization weight) = 667 for iterative kernel-image estimation; 2000 for per-frame and final deconvolution
Controls the trade-off between data fidelity and the gradient prior. The text says these values were pre-validated on a separate validation set, but the final evaluation is on a single image.
mu (kernel sparsity weight) = 0.04
Weights the L1 penalty on the generalized blur kernels in equation (2.7). Pre-validated, but the validation procedure is not detailed.
alpha (hyper-Laplacian exponent) = 0.80
Sets the sparsity level of the gradient prior in equation (2.5). Chosen from the natural-image range [0.5, 0.8] cited from Simoncelli; not fitted to data, but manually fixed.
r1 (manifold neighborhood radius) = 108
Determines which deconvolved frames are averaged in the MF-step. The sensitivity analysis in Figure 8 shows performance peaks near 110, so the reported result uses a value selected on the same test image.
r2 (manifold contraction radius) = 10 times r1
Controls the local contraction weights in equation (2.15). The paper reports the method is not sensitive to r2 for a fixed r1.

assumptions (6)

domain assumption Manifold hypothesis: natural images, and deconvolved RSA frames, lie near a low-dimensional manifold in the ambient pixel space.
Invoked in Section 2.2.2; if false, the MF-step averaging removes detail instead of noise.
domain assumption Imaging model Yi = Ki * X + Ni with i.i.d. Gaussian noise, kernels on the probability simplex, and X independent of Ki.
Equation (2.1) and surrounding text in Section 2; the simulation uses this exact model, so real RSA deviations are untested.
domain assumption Periodic (circular) convolution with padded kernels accurately represents the RSA blur.
Appendix A, equations (A.2) to (A.4); used in all Fourier-domain derivations.
domain assumption Natural image gradients follow a hyper-Laplacian distribution with alpha in [0.5, 0.8].
Section 2.1, equation (2.5); the authors set alpha equal to 0.8.
standard math Half-quadratic splitting solutions converge to the original objective as beta tends to infinity.
Appendix B, equation (B.1); standard result, used to solve the X-step problems.
standard math Convolution theorem and Parseval equality for the 2D discrete Fourier transform.
Appendix A, Lemmas A.1 and A.2; used in Proposition A.1 and Appendix C.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Multi-Frame Blind Manifold Deconvolution for Rotating Synthetic Aperture Imaging." pith.science (2026). https://pith.science/paper/ORJGHGZA

@misc{pith2026250119386,
  author       = {Pith},
  title        = {Pith review of: Multi-Frame Blind Manifold Deconvolution for Rotating Synthetic Aperture Imaging},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ORJGHGZA}},
  note         = {Machine review of arXiv:2501.19386}
}

read the original abstract

Rotating synthetic aperture (RSA) imaging system captures images of the target scene at different rotation angles by rotating a rectangular aperture. Deblurring acquired RSA images plays a critical role in reconstructing a latent sharp image underlying the scene. In the past decade, the emergence of blind convolution technology has revolutionised this field by its ability to model complex features from acquired images. Most of the existing methods attempt to solve the above ill-posed inverse problem through maximising a posterior. Despite this progress, researchers have paid limited attention to exploring low-dimensional manifold structures of the latent image within a high-dimensional ambient-space. Here, we propose a novel method to process RSA images using manifold fitting and penalisation in the content of multi-frame blind convolution. We develop fast algorithms for implementing the proposed procedure. Simulation studies demonstrate that manifold-based deconvolution can outperform conventional deconvolution algorithms in the sense that it can generate a sharper estimate of the latent image in terms of estimating pixel intensities and preserving structural details.

Figures

Figures reproduced from arXiv: 2501.19386 by the authors.

**Figure 2.** First row: 12 estimated blur kernels; second row: the ass [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗

**Figure 3.** First row: 4 deconvolved images; second row: the associa [PITH_FULL_IMAGE:figures/full_fig_p017_3.png] view at source ↗

Figures from the paper (6 more)

**Figure 4.** Figure 4: (a) Comparison of PSNR values of deconvolved images [PITH_FULL_IMAGE:figures/full_fig_p018_4.png]

**Figure 5.** Figure 5: Process of denoising. (a) One blurred image [PITH_FULL_IMAGE:figures/full_fig_p019_5.png]

**Figure 6.** Figure 6: Comparisons of blurred images {yi} n i=1, enhanced images(by applying manifold fitting on the blurred images directly) {y¨i} n i=1’s and enhanced convolved images {yei} n i=1. The baseline images are {Yi} n i=1, i.e., the PSNR (peak signalto-noise ratio) and SSIM (str…

**Figure 7.** Figure 7: Reconstructed sharp images. (a) Zhou et al. (2021) (inp [PITH_FULL_IMAGE:figures/full_fig_p021_7.png]

**Figure 8.** Figure 8: Comparison of the effect of r1 on the latent image reconstruction quality, evaluated using two metrics: PSNR (peak signal-to-noise ratio) and SSIM (structural similarity):(a) Without prior (Equation 2.8). (b) With prior (Equation 2.9, proposed). 22 [PITH_FULL_IMAGE:f…

**Figure 9.** Figure 9: The flowchart of half-quadratic splitting framework. [PITH_FULL_IMAGE:figures/full_fig_p029_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 51 canonical work pages

[1]

Fan, J. (1991). On the optimal rates of convergence for nonp arametric deconvo- lution problems. Ann. Stat. , 19, 1257–1272. 32

work page 1991
[2]

and Freeman, W

Levin, A., Weiss, Y., Durand, F. and Freeman, W. T. (2009). Unde rstanding and evaluating blind deconvolution algorithms. IEEE Trans. Pattern Analysis and Machine Intelligence , 33, 1964-1971

work page 2009
[3]

Full aperture ima ge synthesis using rotating strip aperture image measurements US Patent 5,243,351

Rafanelli, Gerard L and Rehﬁeld, Mark J (1993). Full aperture ima ge synthesis using rotating strip aperture image measurements US Patent 5,243,351

work page 1993
[4]

Fast image deconvolution using hyper- Laplacian priors

Krishnan, Dilip and Fergus, Rob (2009). Fast image deconvolution using hyper- Laplacian priors. Advances in neural information processing systems , 22, 1033– 1041

work page 2009
[5]

Characterization and exp erimental veriﬁcation of the rotating synthetic aperture optical imaging sys tem

Sun, Yu and Zhi, Xiyang and Zhang, Lei and Jiang, Shikai and Shi, T ianjun and Wang, Nan and Gong, Jinnan (2023). Characterization and exp erimental veriﬁcation of the rotating synthetic aperture optical imaging sys tem. Scientiﬁc Reports, 13, 17015

work page 2023
[6]

Low dimen sional mani- fold model for image processing, SIAM Journal on Imaging Sciences , 10, 1669– 1690

Osher, Stanley and Shi, Zuoqiang and Zhu, Wei (2017). Low dimen sional mani- fold model for image processing, SIAM Journal on Imaging Sciences , 10, 1669– 1690

work page 2017
[7]

Unsupervised le arning of image manifolds by semideﬁnite programming, International journal of computer vision, 70, 77–90

Weinberger, Kilian Q and Saul, Lawrence K (2006). Unsupervised le arning of image manifolds by semideﬁnite programming, International journal of computer vision, 70, 77–90

work page 2006
[8]

Manifold Fitting, arXiv preprint arXiv:2304.07680

Yao, Zhigang and Su, Jiaji and Li, Bingjie (2023). Manifold Fitting, arXiv preprint arXiv:2304.07680

arXiv 2023

Show all 54 references

[9]

Bayesian denoising of visual images in the wavelet domain, at Lecture Notes in Statistics , Springer, pp.291-308

Simoncelli, Eero P (1999). Bayesian denoising of visual images in the wavelet domain, at Lecture Notes in Statistics , Springer, pp.291-308. 33

1999
[10]

and Li, Q

Zhou, H., Chen, Y., Feng, H., Lv, G., Xu, Z. and Li, Q. (2021). Rot ated rect- angular aperture imaging through multi-frame blind deconvolution wit h Hyper- Laplacian priors, Optics Express , 29, 12145–12159

2021
[11]

Adam: A method for sto chastic optimization, arXiv preprint arXiv:1412.6980

Kingma, Diederik P and Ba, Jimmy (2014). Adam: A method for sto chastic optimization, arXiv preprint arXiv:1412.6980

2014 arXiv
[12]

and Zhang, L

Zuo, W., Ren, D., Zhang, D., Gu, S. and Zhang, L. (2016). Learn ing iteration- wise generalized shrinkage–thresholding operators for blind decon volution, IEEE Transactions on Image Processing , 25, 1751–1764

2016
[13]

and Ramaswamy, N

Satish, P., Srikantaswamy, M. and Ramaswamy, N. (2020). A Co mprehensive Review of Blind Deconvolution Techniques for Image Deblurring, Traitement du Signal, 37, 527-539

2020
[14]

and Narayanan, H

Feﬀerman, C., Mitter, S. and Narayanan, H. (2016). Testing t he manifold hy- pothesis, Journal of the American Mathematical Society , 29, 983–1049,

2016
[15]

and Medioni, G

Gong, D., Sha, F. and Medioni, G. (2010). Locally linear denoising o n image manifolds, Proceedings of the Thirteenth International Conference on Artiﬁcial Intelligence and Statistics , pp. 265–272,

2010
[16]

and Li, Qi (2012)

Dong, W., Feng, H., Xu, Z. and Li, Qi (2012). Multi-frame blind dec onvolution using sparse priors, Optics Communications , 285, 2276–2288,

2012
[17]

and Velmurugan, R

Rameshan, R., Chaudhuri, S. and Velmurugan, R. (2012). Joint MAP estimation for blind deconvolution: When does it work?, Proceedings of the Eighth Indian Conference on Computer Vision, Graphics and Image Processi ng, pp. 1–7

2012
[18]

and MacKay, D

Miskin, J. and MacKay, D. (2000). Ensemble learning for blind imag e separation and deconvolution, Advances in independent component analysis , pp. 123–141, Springer. 34

2000
[19]

and Lee, S

Cho, S. and Lee, S. (2009). Fast motion deblurring, ACM SIGGRAPH Asia 2009 papers, pp. 1–8,

2009
[20]

and Zhang, D

Zuo, W., Meng, D., Zhang, L., Feng, X. and Zhang, D. (2013). A g eneralized iterated shrinkage algorithm for non-convex sparse coding, Proceedings of the IEEE international conference on computer vision , pp. 217–224

2013
[21]

and Gong, J

Zhi, X., Jiang, S., Zhang, L., Wang, D., Hu, J. and Gong, J. (2021) . Imaging mechanism and degradation characteristic analysis of novel rotat ing synthetic aperture system, Optics and Lasers in Engineering , 139, 106500

2021
[22]

and Gillett, P

Stepp, L., Daggert, L. and Gillett, P. (2003). Estimating the co sts of extremely large telescopes, Future giant telescopes, 4840, 309–321

2003
[23]

and Gong, J

Zhi, X., Jiang, S., Zhang, L., Hu, J., Yu, L., Song, X. and Gong, J. ( 2021). Multi- frame image restoration method for novel rotating synthetic ape rture imaging system, Results in Physics , 23, 103991

2021
[24]

and Zhang, W

Sun, Y., Zhi, X., Jiang, S., Fan, G., Yan, X. and Zhang, W. (2024). Image fusion for the novelty rotating synthetic aperture system based on visio n transformer, Information Fusion, 104, 102163

2024
[25]

A full-aperture image synthes is method for the rotating rectangular aperture system using fourier spec trum restoration, Photonics, 8, 522

Lv, Guomian and Xu, Hao and Feng, Huajun and Xu, Zhihai and Zh ou, Hao and Li, Qi and Chen, Yueting (2021). A full-aperture image synthes is method for the rotating rectangular aperture system using fourier spec trum restoration, Photonics, 8, 522

2021
[26]

Monreal, Benjamin and Rodriguez, Christian and Carney, Ama an d Halliday, Robert and Wang, Mingyuan (2018) Wide Aperture Exoplanet Telesc ope: a low-cost ﬂat conﬁguration for a 100+ meter ground based telesco pe, Journal of Astronomical Telescopes, Instruments, and Systems , 4...

2018
[27]

Testing a modiﬁed PCA-based sharpening approach for ima ge fusion, Remote Sensing , 8, 794

Jel´ enek, Jan and Kopaˇ ckov´ a, Veronika and Kouck´ a, Lucie and Miˇ surec, Jan (2016). Testing a modiﬁed PCA-based sharpening approach for ima ge fusion, Remote Sensing , 8, 794

2016
[28]

Extended dept h-of-ﬁeld imag- ing employing integrated binary phase pupil mask and principal compo nent anal- ysis image fusion, Optics Express , 28, 23862–23873

Milgrom, Benjamin and Avrahamy, Roy and David, Tal and Caspi, A vi and Golovachev, Yosef and Engelberg, Shlomo (2020). Extended dept h-of-ﬁeld imag- ing employing integrated binary phase pupil mask and principal compo nent anal- ysis image fusion, Optics Express , 28, 23862–23873

2020
[29]

Principal components colour display of ERTS imagery, NASA

Taylor, MM (1974). Principal components colour display of ERTS imagery, NASA. Goddard Space Flight Center 3d ERTS-1 Symp., Vol. 1, Se ct. B, PAPER- I12

1974
[30]

Extracting spectral con trast in Landsat Thematic Mapper image data using selective principal component ana lysis, Pho- togramm

Kwarteng, P and Chavez, A (1989). Extracting spectral con trast in Landsat Thematic Mapper image data using selective principal component ana lysis, Pho- togramm. Eng. Remote Sens , 55, 339–348

1989
[31]

Remote sensing image fusion via wavelet transform and spa rse represen- tation, ISPRS journal of photogrammetry and remote sensing , 104, 158–173

Cheng, Jian and Liu, Haijun and Liu, Ting and Wang, Feng and Li, Ho ngsheng (2015). Remote sensing image fusion via wavelet transform and spa rse represen- tation, ISPRS journal of photogrammetry and remote sensing , 104, 158–173

2015
[32]

Multisens or image fusion using the wavelet transform, Graphical models and image processing , 57, 235–245

Li, Hui and Manjunath, BS and Mitra, Sanjit K (1995). Multisens or image fusion using the wavelet transform, Graphical models and image processing , 57, 235–245

1995
[33]

Wavele t based image fusion techniques—An introduction, review and comparison, ISPRS Journal of photogrammetry and Remote Sensing , 62, 249–263

Amolins, Krista and Zhang, Yun and Dare, Peter (2007). Wavele t based image fusion techniques—An introduction, review and comparison, ISPRS Journal of photogrammetry and Remote Sensing , 62, 249–263. 36

2007
[34]

Proper image subtraction—optimal transient detection, photometry, and hyp othesis testing, The Astrophysical Journal , 830, 27

Zackay, Barak and Ofek, Eran O and Gal-Yam, Avishay (2016). Proper image subtraction—optimal transient detection, photometry, and hyp othesis testing, The Astrophysical Journal , 830, 27

2016
[35]

Stochastic deconvolution, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pp

Gregson, James and Heide, Felix and Hullin, Matthias B and Rouf, M ushﬁqur and Heidrich, Wolfgang (2013). Stochastic deconvolution, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pp. 1043–1050

2013
[36]

Vari- ational Bayesian blind deconvolution using a total variation prior, IEEE Trans- actions on Image Processing , 18, 12–26

Babacan, S Derin and Molina, Rafael and Katsaggelos, Aggelos K (2008). Vari- ational Bayesian blind deconvolution using a total variation prior, IEEE Trans- actions on Image Processing , 18, 12–26

2008
[37]

Revisiting Bayesian blind d econvolu- tion, Journal of Machine Learning Research , 15, 3775–3814

Wipf, David and Zhang, Haichao (2014). Revisiting Bayesian blind d econvolu- tion, Journal of Machine Learning Research , 15, 3775–3814

2014
[38]

A piecewise local regularized Richardson–Lucy algorithm for remote sensing ima ge deconvo- lution, Optics & Laser Technology , 43, 926–933

Dong, Wende and Feng, Huajun and Xu, Zhihai and Li, Qi (2011) . A piecewise local regularized Richardson–Lucy algorithm for remote sensing ima ge deconvo- lution, Optics & Laser Technology , 43, 926–933

2011
[39]

Schneider, Dorian and van Ekeris, Tilo and zur Jacobsmuehlen, J oschka and Gross, Sebastian (2013). On benchmarking non-blind deconvolutio n algorithms: a sample driven comparison of image de-blurring methods for automa ted vi- sual inspection systems, IEEE International Instr...

2013
[40]

Kotera, Jan and ˇSroubek, Filip and Milanfar, Peyman (2013). Blind decon- volution using alternating maximum a posteriori estimation with heavy -tailed priors, Computer Analysis of Images and Patterns: 15th Internation al Confer- ence, CAIP 2013, York, UK, August 27-29, 2013, ...

2013
[41]

Robust multichanne l blind decon- volution via fast alternating minimization, IEEE Transactions on Image pro- cessing, 21, 1687–1700

Sroubek, Filip and Milanfar, Peyman (2011). Robust multichanne l blind decon- volution via fast alternating minimization, IEEE Transactions on Image pro- cessing, 21, 1687–1700

2011
[42]

l 0- regularized intensity and gradient prior for deblurring text images a nd beyond, IEEE transactions on pattern analysis and machine intellig ence, 39, 342–355

Pan, Jinshan and Hu, Zhe and Su, Zhixun and Yang, Ming-Hsuan ( 2016). l 0- regularized intensity and gradient prior for deblurring text images a nd beyond, IEEE transactions on pattern analysis and machine intellig ence, 39, 342–355

2016
[43]

Blind d econvolution using a normalized sparsity measure, CVPR 2011 , 233–240

Krishnan, Dilip and Tay, Terence and Fergus, Rob (2011). Blind d econvolution using a normalized sparsity measure, CVPR 2011 , 233–240

2011
[44]

Weighted nuclear norm minimization with application to image denoising, Pro- ceedings of the IEEE conference on computer vision and patte rn recognition, pp

Gu, Shuhang and Zhang, Lei and Zuo, Wangmeng and Feng, Xian gchu (2014). Weighted nuclear norm minimization with application to image denoising, Pro- ceedings of the IEEE conference on computer vision and patte rn recognition, pp. 2862–2869

2014
[45]

Single image ha ze removal using dark channel prior, IEEE transactions on pattern analysis and machine intelligence, 33, 2341–2353

He, Kaiming and Sun, Jian and Tang, Xiaoou (2010). Single image ha ze removal using dark channel prior, IEEE transactions on pattern analysis and machine intelligence, 33, 2341–2353

2010
[46]

Image deblurring via enha nced low- rank prior, IEEE Transactions on Image Processing , 25, 3426–3437

Ren, Wenqi and Cao, Xiaochun and Pan, Jinshan and Guo, Xiaojie and Zuo, Wangmeng and Yang, Ming-Hsuan (2016). Image deblurring via enha nced low- rank prior, IEEE Transactions on Image Processing , 25, 3426–3437

2016
[47]

Blind image de blurring using spectral properties of convolution operators, IEEE Transactions on image processing, 23, 5047–5056

Liu, Guangcan and Chang, Shiyu and Ma, Yi (2014). Blind image de blurring using spectral properties of convolution operators, IEEE Transactions on image processing, 23, 5047–5056

2014
[48]

Total variation blin d deconvolu- tion, IEEE transactions on Image Processing , 7, 370–375

Chan, Tony F and Wong, Chiu-Kwong (1998). Total variation blin d deconvolu- tion, IEEE transactions on Image Processing , 7, 370–375. 38

1998
[49]

Removing camer a shake via weighted fourier burst accumulation, IEEE Transactions on Image Processing , 24, 3293–3307

Delbracio, Mauricio and Sapiro, Guillermo (2015). Removing camer a shake via weighted fourier burst accumulation, IEEE Transactions on Image Processing , 24, 3293–3307

2015
[50]

Neural blind deconvolution using deep priors, Proceedings of the IEEE/CVF conference on computer vision and pattern re cognition, pp

Ren, Dongwei and Zhang, Kai and Wang, Qilong and Hu, Qinghua a nd Zuo, Wangmeng (2020). Neural blind deconvolution using deep priors, Proceedings of the IEEE/CVF conference on computer vision and pattern re cognition, pp. 3341–3350

2020
[51]

1954–1958

Kotera, Jan and ˇSroubek, Filip and ˇSm´ιdl, V´ aclav (2021).Improving neural blind deconvolution, IEEE International Conference on Image Processing (ICIP) , pp. 1954–1958

2021
[52]

Celer: a Fast Solver for the Lasso with Dual Extrapolation, Proceedings of the 35th International Conference on Machine Learning , 80, 3321–3330

Massias, Mathurin and Gramfort, Alexandre and Salmon, Josep h (2018). Celer: a Fast Solver for the Lasso with Dual Extrapolation, Proceedings of the 35th International Conference on Machine Learning , 80, 3321–3330

2018
[53]

Dual Extrapolation for Sparse GLMs , Journal of Machine Learn- ing Research, 21, 1-33

Mathurin Massias and Samuel Vaiter and Alexandre Gramfort an d Joseph Salmon (2020). Dual Extrapolation for Sparse GLMs , Journal of Machine Learn- ing Research, 21, 1-33

2020
[54]

Proje ction onto the probability simplex: An eﬃcient algorithm with a simple proof, and an app lica- tion, arXiv preprint arXiv:1309.1541

Wang, Weiran and Carreira-Perpin´ an, Miguel A (2013). Proje ction onto the probability simplex: An eﬃcient algorithm with a simple proof, and an app lica- tion, arXiv preprint arXiv:1309.1541 . 39

2013 arXiv

Pith tools

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

[1] [1]

Fan, J. (1991). On the optimal rates of convergence for nonp arametric deconvo- lution problems. Ann. Stat. , 19, 1257–1272. 32

work page 1991

[2] [2]

and Freeman, W

Levin, A., Weiss, Y., Durand, F. and Freeman, W. T. (2009). Unde rstanding and evaluating blind deconvolution algorithms. IEEE Trans. Pattern Analysis and Machine Intelligence , 33, 1964-1971

work page 2009

[3] [3]

Full aperture ima ge synthesis using rotating strip aperture image measurements US Patent 5,243,351

Rafanelli, Gerard L and Rehﬁeld, Mark J (1993). Full aperture ima ge synthesis using rotating strip aperture image measurements US Patent 5,243,351

work page 1993

[4] [4]

Fast image deconvolution using hyper- Laplacian priors

Krishnan, Dilip and Fergus, Rob (2009). Fast image deconvolution using hyper- Laplacian priors. Advances in neural information processing systems , 22, 1033– 1041

work page 2009

[5] [5]

Characterization and exp erimental veriﬁcation of the rotating synthetic aperture optical imaging sys tem

Sun, Yu and Zhi, Xiyang and Zhang, Lei and Jiang, Shikai and Shi, T ianjun and Wang, Nan and Gong, Jinnan (2023). Characterization and exp erimental veriﬁcation of the rotating synthetic aperture optical imaging sys tem. Scientiﬁc Reports, 13, 17015

work page 2023

[6] [6]

Low dimen sional mani- fold model for image processing, SIAM Journal on Imaging Sciences , 10, 1669– 1690

Osher, Stanley and Shi, Zuoqiang and Zhu, Wei (2017). Low dimen sional mani- fold model for image processing, SIAM Journal on Imaging Sciences , 10, 1669– 1690

work page 2017

[7] [7]

Unsupervised le arning of image manifolds by semideﬁnite programming, International journal of computer vision, 70, 77–90

Weinberger, Kilian Q and Saul, Lawrence K (2006). Unsupervised le arning of image manifolds by semideﬁnite programming, International journal of computer vision, 70, 77–90

work page 2006

[8] [8]

Manifold Fitting, arXiv preprint arXiv:2304.07680

Yao, Zhigang and Su, Jiaji and Li, Bingjie (2023). Manifold Fitting, arXiv preprint arXiv:2304.07680

arXiv 2023