pith. sign in

arxiv: 2605.23282 · v1 · pith:JCCFMMK5new · submitted 2026-05-22 · 📡 eess.IV · cs.CV· cs.LG

Discontinuous Galerkin Neural Operator for Pathology Defocus Deblurring

Shaoqing Duan , Haofei Song , Xintian Mao , Qingli Li , Yan Wang This is my paper

Pith reviewed 2026-05-25 03:07 UTC · model grok-4.3

classification 📡 eess.IV cs.CVcs.LG
keywords defocus deblurringneural operatordiscontinuous Galerkinpathology microscopyimage restorationspatially varying blurintegral operatormicroscopy imaging
0
0 comments X

The pith

A discontinuous Galerkin neural operator models position-dependent defocus blur as an integral operator for pathology image deblurring.

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

The paper aims to show that existing neural networks for image deblurring fail on pathology microscopy because they assume uniform blur across the image, while real defocus varies by position and can be discontinuous. It proposes instead to treat the blur as a learnable integral operator whose kernel is built from local discontinuous Galerkin elements and numerical fluxes at interfaces. This preserves the physics of light collection while allowing the network to adapt to heterogeneous blur. If successful, the method would produce sharper, more accurate reconstructions of tissue samples at high resolution without the limitations of shift-invariant assumptions.

Core claim

The Discontinuous Galerkin Neural Operator (DGNO) parameterizes the integral kernel of defocus formation using element-local volume operators and interface numerical fluxes. This formulation combines locality, heterogeneity modeling, and global coherence while preserving the underlying physics of optical image formation. Extensive experiments show it surpasses state-of-the-art methods in delivering sharper reconstructions, robust handling of spatially varying blur, and scalable high-resolution performance.

What carries the argument

The Discontinuous Galerkin Neural Operator, which parameterizes the defocus integral kernel using a discontinuous Galerkin formulation with element-local volume operators and interface numerical fluxes.

If this is right

  • Sharper reconstructions of pathology images with spatially varying blur become possible.
  • Scalable performance on high-resolution images is achieved without assuming shift-invariance.
  • Direct modeling of the integral imaging process provides better interpretability than black-box networks.
  • The approach maintains physical consistency in the deblurring process.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This formulation might generalize to other inverse problems in imaging where blur is locally discontinuous.
  • Integrating DGNO with multi-scale architectures could further improve handling of varying blur scales.
  • Real-time clinical deployment would require testing inference speed on standard hardware.

Load-bearing premise

Defocus formation in pathology microscopy can be directly modeled as an integral operator whose kernel is usefully parameterized by a discontinuous Galerkin formulation.

What would settle it

A dataset of pathology images with known spatially varying defocus where DGNO fails to produce sharper results than standard convolutional networks would falsify the superiority claim.

Figures

Figures reproduced from arXiv: 2605.23282 by Haofei Song, Qingli Li, Shaoqing Duan, Xintian Mao, Yan Wang.

Figure 1
Figure 1. Figure 1: Comparisons Our DGNO and other state-of-the-art algorithms. Performance and parameters on BBBC006w1 (left) and FLOPs (right) for defocus deblurring. degrade cellular morphology and compromise downstream pathological analysis (Zhang et al., 2022), including cell detection (Schmidt et al., 2018) and segmentation (Keaton et al., 2023). Many image defocus deblurring networks have been proposed. However, when t… view at source ↗
Figure 2
Figure 2. Figure 2: Overview of defocus-blur formation and operator￾learning approaches. (a) Defocus-Blur Formation. (b) SRNO versus the proposed DGNO. spatially varying integral operator: g(x, y) = Z Z K(x, y; ξ, η) h(ξ, η) dξdη, (1) where K denotes the PSF determined by the optical system. Only under the restrictive assumption of shift invariance does this formulation reduce to a standard convolution. But, this assumption i… view at source ↗
Figure 3
Figure 3. Figure 3: The proposed Discontinuous Galerkin Neural Operator (DGNO) architecture for defocus deblurring by lifting a defocus image x(r) into a feature space using a mamba encoder. Kernel integrals composed of T layers of discontinuous attention. This pipeline generates s(r), a sharp images of the input image. integral operator and does not contain differential terms, so it cannot be directly decomposed by integrati… view at source ↗
Figure 4
Figure 4. Figure 4: Visual comparison of single image defocus deblur approaches on BBBC006w1 and BBBC006w2. channel dimensions [48, 96, 192, 384]. DGNO operates on non-overlapping local elements of size 8×8. The model was trained on NVIDIA RTX4090 GPUs (48 GB) using AdamW optimizer (β1 = 0.9, β2 = 0.999, weight decay=1 × 10−4 ) with an initial learning rate of 3 × 10−4 (cosine decay to 1 × 10−6 ). The batch size is 8 with tra… view at source ↗
Figure 6
Figure 6. Figure 6: Residual visualization and edge-aware quantitative anal￾ysis comparing GG and DG operators. (a) Blur image (b) Sharp image. (c) Absolute error difference |eGG| − |eDG|. (d) PSNR comparison on edge-band and non-edge regions [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗
Figure 5
Figure 5. Figure 5: Effective-rank analysis of latent representations across scales and iteration steps (T = 0, 1). (a) the absolute effective rank of the latent matrix z, (b) the effective-rank utilization reff /d To validate the discontinuous pattern learning ability of DGNO, as shown in [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: LAM-based interpretability and defocus deblurring re￾sults on BBBC006. aligned activations produced by DGNO (highlighted by the blue arrows in [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 9
Figure 9. Figure 9: Visual comparison with non-blind spatially varying deconvolution on BBBC006w1 under Seidel-coefficient-based blur. Numerical Flux and Boundary Condition Analysis. Ta￾ble 3 presents an ablation study of numerical fluxes and boundary conditions for DGNO-Face and DGNO-Cell on BBBC006w1. For DGNO-Face, the average-jump flux under Dirichlet boundary conditions achieves the highest PSNR, whereas for DGNO-Cell, t… view at source ↗
Figure 10
Figure 10. Figure 10: Visualization of the learned dynamic basis functions, latent representations Q(K⊤V ), and DG flux and boundary behavior on BBBC006 (Ljosa et al., 2012). F. Dataset Description. BBBC006 (Ljosa et al., 2012): This dataset from the Broad Bioimage Benchmark Collection (BBBC) consists of fluorescence microscopy images in two channels, denoted as w1 (Hoechst-stained nuclei) and w2 (Phalloidin-stained actin). Fo… view at source ↗
Figure 11
Figure 11. Figure 11: Visual comparison of single image defocus deblur approaches on BBBC006w1 (Ljosa et al., 2012) 16 [PITH_FULL_IMAGE:figures/full_fig_p016_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Visual comparison of single image defocus deblur approaches on BBBC006w2 (Ljosa et al., 2012) (a) Blur (b) GKMNet (c) MIMOUNet (d) SwinIR (e) MambaIRv2 (f) Restormer (g) MPT+EFCR (h) DGNO(face) (i) DGNO(cell) (j) Sharp [PITH_FULL_IMAGE:figures/full_fig_p017_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Visual comparison of single image defocus deblur approaches on 3DHistech (Geng et al., 2022) 17 [PITH_FULL_IMAGE:figures/full_fig_p017_13.png] view at source ↗
read the original abstract

Defocus deblurring in pathological microscopy remains challenging due to the spatially varying and locally discontinuous nature of optical blur induced by a position-dependent integral imaging process. Existing deep learning methods, constrained by shift-invariance assumptions and limited interpretability, are not well suited to such heterogeneous blur patterns. Neural operators provide a principled alternative by modeling defocus formation directly as an integral operator, offering a new perspective on defocus deblurring. However, most existing neural operator architectures for low-level vision rely on globally parameterized kernels that assume smoothness and stationarity, limiting their ability to model heterogeneous and locally discontinuous blur patterns. To address this limitation, we propose the Discontinuous Galerkin Neural Operator (DGNO), which parameterizes the integral kernel using a discontinuous Galerkin formulation with element-local volume operators and interface numerical fluxes. DGNO provides a principled combination of locality, heterogeneity modeling, and global coherence while preserving the underlying physics of optical image formation. Extensive and insightful experiments demonstrate that DGNO surpasses state-of-the-arts, delivering sharper reconstructions, robust handling of spatially varying blur, and scalable high-resolution performance. The code will be released at https://github.com/DeepMed-Lab-ECNU/Single-Image-Deblur.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The manuscript proposes the Discontinuous Galerkin Neural Operator (DGNO) to address defocus deblurring in pathological microscopy. It frames defocus formation as a position-dependent integral operator and parameterizes the kernel via a discontinuous Galerkin discretization that uses element-local volume operators and interface numerical fluxes, aiming to capture spatial heterogeneity and local discontinuities while preserving optical physics. The central claim is that this architecture outperforms existing shift-invariant deep learning methods and other neural operators on sharpness, robustness to varying blur, and scalability to high resolution, with code to be released.

Significance. If the performance claims are substantiated, the work would supply a physics-aligned neural operator architecture that explicitly targets spatially varying blur, a common limitation in microscopy. The explicit release of code is a positive step toward reproducibility. The modeling choice of combining locality, heterogeneity, and global coherence via DG elements is a coherent extension of neural operators to low-level vision tasks with discontinuous kernels.

major comments (1)
  1. Abstract: the assertion that 'extensive and insightful experiments demonstrate that DGNO surpasses state-of-the-arts' is unsupported by any quantitative metrics, tables, error bars, dataset descriptions, or ablation results in the supplied manuscript text; this directly undermines evaluation of the central performance claim.
minor comments (1)
  1. The abstract mentions 'the code will be released' but provides no link or repository details beyond the GitHub path; this should be confirmed in the final version.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their review and for highlighting this issue with the abstract. We address the comment point-by-point below.

read point-by-point responses

  1. Referee: Abstract: the assertion that 'extensive and insightful experiments demonstrate that DGNO surpasses state-of-the-arts' is unsupported by any quantitative metrics, tables, error bars, dataset descriptions, or ablation results in the supplied manuscript text; this directly undermines evaluation of the central performance claim.

    Authors: We agree that the abstract's performance claim would be stronger if accompanied by concrete quantitative indicators. The full manuscript contains dedicated experimental sections with tables, error bars, dataset details, and ablations; however, these details are not summarized in the abstract itself. We will revise the abstract to incorporate key quantitative results (e.g., PSNR/SSIM improvements and dataset sizes) so that the claim is directly supported within the abstract text. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper introduces DGNO as an architectural modeling choice that parameterizes an integral operator for defocus deblurring using discontinuous Galerkin elements with local volume operators and interface fluxes. This is framed as a direct response to the shift-invariance limitations of prior neural operators, without any equations or steps that reduce the claimed performance gains to fitted inputs, self-definitions, or self-citation chains by construction. The derivation chain consists of a coherent proposal plus experimental validation, remaining self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central modeling choice rests on treating optical blur as an integral operator amenable to discontinuous Galerkin discretization; no free parameters, invented entities, or additional axioms are stated in the abstract.

axioms (1)
  • domain assumption Defocus formation is an integral operator that can be discretized with discontinuous Galerkin elements and numerical fluxes
    Directly stated in the abstract as the principled alternative to shift-invariant assumptions.

pith-pipeline@v0.9.0 · 5769 in / 1076 out tokens · 19873 ms · 2026-05-25T03:07:42.010878+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

96 extracted references · 96 canonical work pages · 6 internal anchors

  1. [1]

    Proceedings of the IEEE/CVF international conference on computer vision , pages=

    Swin transformer: Hierarchical vision transformer using shifted windows , author=. Proceedings of the IEEE/CVF international conference on computer vision , pages=

    work page

  2. [2]

    Advances in Neural Information Processing Systems , volume=

    Gaussian kernel mixture network for single image defocus deblurring , author=. Advances in Neural Information Processing Systems , volume=

    work page

  3. [3]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Restormer: Efficient transformer for high-resolution image restoration , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    work page

  4. [4]

    Proceedings of the IEEE/CVF international conference on computer vision , pages=

    Rethinking coarse-to-fine approach in single image deblurring , author=. Proceedings of the IEEE/CVF international conference on computer vision , pages=

    work page

  5. [5]

    European conference on computer vision , pages=

    Simple baselines for image restoration , author=. European conference on computer vision , pages=. 2022 , organization=

    work page 2022

  6. [6]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    A unified framework for microscopy defocus deblur with multi-pyramid transformer and contrastive learning , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    work page

  7. [7]

    arXiv preprint arXiv:2503.12399 , year=

    Pathology Image Restoration via Mixture of Prompts , author=. arXiv preprint arXiv:2503.12399 , year=

    work page arXiv

  8. [8]

    European conference on computer vision , pages=

    Mambair: A simple baseline for image restoration with state-space model , author=. European conference on computer vision , pages=. 2024 , organization=

    work page 2024

  9. [9]

    Proceedings of the 32nd ACM International Conference on Multimedia , pages=

    Loformer: Local frequency transformer for image deblurring , author=. Proceedings of the 32nd ACM International Conference on Multimedia , pages=

    work page

  10. [10]

    Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    Building Vision Models upon Heat Conduction , author=. Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    work page

  11. [11]

    Nature methods , volume=

    Annotated high-throughput microscopy image sets for validation , author=. Nature methods , volume=

    work page

  12. [12]

    Medical Image Analysis , volume=

    Cervical cytopathology image refocusing via multi-scale attention features and domain normalization , author=. Medical Image Analysis , volume=. 2022 , publisher=

    work page 2022

  13. [13]

    arXiv preprint arXiv:2211.13228 , year=

    Self-supervised learning based on heat equation , author=. arXiv preprint arXiv:2211.13228 , year=

    work page arXiv

  14. [14]

    Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    Object detection using event camera: A moe heat conduction based detector and a new benchmark dataset , author=. Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    work page

  15. [15]

    arXiv preprint arXiv:2505.12908 , year=

    Dynamic Graph Induced Contour-aware Heat Conduction Network for Event-based Object Detection , author=. arXiv preprint arXiv:2505.12908 , year=

    work page arXiv

  16. [16]

    arXiv preprint arXiv:2504.05830 , year=

    Human Activity Recognition using RGB-Event based Sensors: A Multi-modal Heat Conduction Model and A Benchmark Dataset , author=. arXiv preprint arXiv:2504.05830 , year=

    work page arXiv

  17. [17]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    RS-vHeat: Heat Conduction Guided Efficient Remote Sensing Foundation Model , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    work page

  18. [18]

    1976 , publisher=

    The heat equation , author=. 1976 , publisher=

    work page 1976

  19. [19]

    Biophysics Reviews , volume=

    Live cell microscopy: From image to insight , author=. Biophysics Reviews , volume=. 2022 , publisher=

    work page 2022

  20. [20]

    Journal of biomedical optics , volume=

    Comprehensive review of surgical microscopes: technology development and medical applications , author=. Journal of biomedical optics , volume=. 2021 , publisher=

    work page 2021

  21. [21]

    Encyclopedia of Life Sciences , volume=

    History of the optical microscope in cell biology and medicine , author=. Encyclopedia of Life Sciences , volume=. 2008 , publisher=

    work page 2008

  22. [22]

    2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI) , pages=

    Deblurring cataract surgery videos using a multi-scale deconvolutional neural network , author=. 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI) , pages=. 2020 , organization=

    work page 2020

  23. [23]

    arXiv preprint arXiv:2011.11879 , year=

    Blind deblurring for microscopic pathology images using deep learning networks , author=. arXiv preprint arXiv:2011.11879 , year=

    work page arXiv 2011

  24. [24]

    2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI) , pages=

    Deep learning framework for mobile microscopy , author=. 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI) , pages=. 2021 , organization=

    work page 2021

  25. [25]

    International Conference on Computer Analysis of Images and Patterns , pages=

    Defocus Blur Synthesis and Deblurring via Interpolation and Extrapolation in Latent Space , author=. International Conference on Computer Analysis of Images and Patterns , pages=. 2023 , organization=

    work page 2023

  26. [26]

    2022 IEEE International Conference on Image Processing (ICIP) , pages=

    Defocus deblur microscopy via head-to-tail cross-scale fusion , author=. 2022 IEEE International Conference on Image Processing (ICIP) , pages=. 2022 , organization=

    work page 2022

  27. [27]

    European conference on computer vision , pages=

    Defocus deblurring using dual-pixel data , author=. European conference on computer vision , pages=. 2020 , organization=

    work page 2020

  28. [28]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Iterative filter adaptive network for single image defocus deblurring , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    work page

  29. [29]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Deep defocus map estimation using domain adaptation , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    work page

  30. [30]

    IEEE Transactions on Image Processing , volume=

    Defocus image deblurring network with defocus map estimation as auxiliary task , author=. IEEE Transactions on Image Processing , volume=. 2021 , publisher=

    work page 2021

  31. [31]

    IEEE Transactions on Computational Imaging , volume=

    Aifnet: All-in-focus image restoration network using a light field-based dataset , author=. IEEE Transactions on Computational Imaging , volume=. 2021 , publisher=

    work page 2021

  32. [32]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Neumann network with recursive kernels for single image defocus deblurring , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    work page

  33. [33]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Learning to deblur using light field generated and real defocus images , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    work page

  34. [34]

    Proceedings of the IEEE/CVF international conference on computer vision , pages=

    Single image defocus deblurring using kernel-sharing parallel atrous convolutions , author=. Proceedings of the IEEE/CVF international conference on computer vision , pages=

    work page

  35. [35]

    Precision Engineering , volume=

    Deblurring microscopic image by integrated convolutional neural network , author=. Precision Engineering , volume=. 2023 , publisher=

    work page 2023

  36. [36]

    Computational and Structural Biotechnology Journal , volume=

    Correction of out-of-focus microscopic images by deep learning , author=. Computational and Structural Biotechnology Journal , volume=. 2022 , publisher=

    work page 2022

  37. [37]

    Advances in neural information processing systems , volume=

    Attention is all you need , author=. Advances in neural information processing systems , volume=

    work page

  38. [38]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Efficient and explicit modelling of image hierarchies for image restoration , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    work page

  39. [39]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Adarevd: Adaptive patch exiting reversible decoder pushes the limit of image deblurring , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    work page

  40. [40]

    Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    Quad-Pixel Image Defocus Deblurring: A New Benchmark and Model , author=. Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    work page

  41. [41]

    Advances in Neural Information Processing Systems , volume=

    On the parameterization and initialization of diagonal state space models , author=. Advances in Neural Information Processing Systems , volume=

    work page

  42. [42]

    Advances in neural information processing systems , volume=

    S4nd: Modeling images and videos as multidimensional signals with state spaces , author=. Advances in neural information processing systems , volume=

    work page

  43. [43]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Selective structured state-spaces for long-form video understanding , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    work page

  44. [44]

    Retentive Network: A Successor to Transformer for Large Language Models

    Retentive network: A successor to transformer for large language models , author=. arXiv preprint arXiv:2307.08621 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  45. [45]

    RWKV: Reinventing RNNs for the Transformer Era

    Rwkv: Reinventing rnns for the transformer era , author=. arXiv preprint arXiv:2305.13048 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  46. [46]

    Advances in neural information processing systems , volume=

    Denoising diffusion probabilistic models , author=. Advances in neural information processing systems , volume=

    work page

  47. [47]

    Advances in neural information processing systems , volume=

    Photorealistic text-to-image diffusion models with deep language understanding , author=. Advances in neural information processing systems , volume=

    work page

  48. [48]

    Denoising Diffusion Implicit Models

    Denoising diffusion implicit models , author=. arXiv preprint arXiv:2010.02502 , year=

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  49. [49]

    Materials Today: Proceedings , volume=

    Object and character recognition using spiking neural network , author=. Materials Today: Proceedings , volume=. 2018 , publisher=

    work page 2018

  50. [50]

    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

    An image is worth 16x16 words: Transformers for image recognition at scale , author=. arXiv preprint arXiv:2010.11929 , year=

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  51. [51]

    Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , pages=

    Improving single-image defocus deblurring: How dual-pixel images help through multi-task learning , author=. Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , pages=

    work page

  52. [52]

    European Conference on Computer Vision , pages=

    United defocus blur detection and deblurring via adversarial promoting learning , author=. European Conference on Computer Vision , pages=. 2022 , organization=

    work page 2022

  53. [53]

    arXiv preprint arXiv:2508.12736 , year=

    Frequency-Driven Inverse Kernel Prediction for Single Image Defocus Deblurring , author=. arXiv preprint arXiv:2508.12736 , year=

    work page arXiv

  54. [54]

    IEEE Transactions on Pattern Analysis and Machine Intelligence , year=

    Deep single image defocus deblurring via gaussian kernel mixture learning , author=. IEEE Transactions on Pattern Analysis and Machine Intelligence , year=

    work page

  55. [55]

    Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    Performing defocus deblurring by modeling its formation process , author=. Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=

    work page

  56. [56]

    International Journal of Computer Vision , volume=

    Pyramid attention network for image restoration , author=. International Journal of Computer Vision , volume=. 2023 , publisher=

    work page 2023

  57. [57]

    European conference on computer vision , pages=

    Mssnet: Multi-scale-stage network for single image deblurring , author=. European conference on computer vision , pages=. 2022 , organization=

    work page 2022

  58. [58]

    Proceedings of the IEEE/CVF international conference on computer vision , pages=

    Single image defocus deblurring via implicit neural inverse kernels , author=. Proceedings of the IEEE/CVF international conference on computer vision , pages=

    work page

  59. [59]

    Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    Parameterized Blur Kernel Prior Learning for Local Motion Deblurring , author=. Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    work page

  60. [60]

    Advances in neural information processing systems , volume=

    Vmamba: Visual state space model , author=. Advances in neural information processing systems , volume=

    work page

  61. [61]

    Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model

    Vision mamba: Efficient visual representation learning with bidirectional state space model , author=. arXiv preprint arXiv:2401.09417 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  62. [62]

    arXiv preprint arXiv:2403.08330 , year=

    Activating wider areas in image super-resolution , author=. arXiv preprint arXiv:2403.08330 , year=

    work page arXiv

  63. [63]

    IEEE Transactions on Circuits and Systems for Video Technology , year=

    Vmambair: Visual state space model for image restoration , author=. IEEE Transactions on Circuits and Systems for Video Technology , year=

    work page

  64. [64]

    2024 IEEE 7th International Conference on Multimedia Information Processing and Retrieval (MIPR) , pages=

    Cu-mamba: Selective state space models with channel learning for image restoration , author=. 2024 IEEE 7th International Conference on Multimedia Information Processing and Retrieval (MIPR) , pages=. 2024 , organization=

    work page 2024

  65. [65]

    Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    Mambairv2: Attentive state space restoration , author=. Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    work page

  66. [66]

    International Journal of Computer Vision , pages=

    Reblurring-Guided Single Image Defocus Deblurring: A Learning Framework with Misaligned Training Pairs , author=. International Journal of Computer Vision , pages=. 2025 , publisher=

    work page 2025

  67. [67]

    Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages=

    Just noticeable defocus blur detection and estimation , author=. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages=

    work page

  68. [68]

    Proceedings of the IEEE international conference on computer vision , pages=

    Estimating defocus blur via rank of local patches , author=. Proceedings of the IEEE international conference on computer vision , pages=

    work page

  69. [69]

    Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    A unified approach of multi-scale deep and hand-crafted features for defocus estimation , author=. Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    work page

  70. [70]

    IEEE Transactions on Image Processing , volume=

    Edge-based defocus blur estimation with adaptive scale selection , author=. IEEE Transactions on Image Processing , volume=. 2017 , publisher=

    work page 2017

  71. [71]

    IEEE Transactions on Image Processing , volume=

    Non-parametric blur map regression for depth of field extension , author=. IEEE Transactions on Image Processing , volume=. 2016 , publisher=

    work page 2016

  72. [72]

    ACM SIGGRAPH 2007 papers , pages=

    Image deblurring with blurred/noisy image pairs , author=. ACM SIGGRAPH 2007 papers , pages=

    work page 2007

  73. [73]

    IEEE transactions on pattern analysis and machine intelligence , volume=

    Defocus blur detection via multi-stream bottom-top-bottom network , author=. IEEE transactions on pattern analysis and machine intelligence , volume=. 2019 , publisher=

    work page 2019

  74. [74]

    Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    Deep learning for handling kernel/model uncertainty in image deconvolution , author=. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages=

    work page

  75. [75]

    Advances in neural information processing systems , volume=

    Deep non-blind deconvolution via generalized low-rank approximation , author=. Advances in neural information processing systems , volume=

    work page

  76. [76]

    Acm Transactions on Graphics (TOG) , volume=

    Progressive inter-scale and intra-scale non-blind image deconvolution , author=. Acm Transactions on Graphics (TOG) , volume=. 2008 , publisher=

    work page 2008

  77. [77]

    Fourier Neural Operator for Parametric Partial Differential Equations

    Fourier neural operator for parametric partial differential equations , author=. arXiv preprint arXiv:2010.08895 , year=

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  78. [78]

    Journal of Machine Learning Research , volume=

    Neural operator: Learning maps between function spaces with applications to pdes , author=. Journal of Machine Learning Research , volume=

    work page

  79. [79]

    Nature machine intelligence , volume=

    Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators , author=. Nature machine intelligence , volume=. 2021 , publisher=

    work page 2021

  80. [80]

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    Super-resolution neural operator , author=. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=

    work page

Showing first 80 references.