Pith. sign in

REVIEW 3 major objections 2 minor 53 references

Wavelet-Guided Dual-Frequency Encoding for Remote Sensing Change Detection

T0 review · 3 major / 2 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Wavelet-based dual-frequency encoding sharpens edge and subtle-change detection in remote sensing imagery.

desk verdict The architecture is new, but the supplied full text is unreadable — no referee can verify the central accuracy claim. read the letter →

arxiv 2508.05271 v1 pith:ETAL34GK submitted 2025-08-07 cs.CV

Xiaoyang Zhang , Guodong Fan , Guang-Yong Chen , Zhen Hua , Jinjiang Li , Min Gan , C. L. Philip Chen This is my paper · ORCID
classification cs.CV
keywords changedetectionremotesensingdiscretewavelettransformdual-frequencyencodingedgeambiguitytransformerdeeplearning
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

The paper tries to show that representing remote-sensing image pairs in the wavelet frequency domain, rather than only in the spatial domain, makes change detection better at exactly the cases that usually fail: thin edges and subtle, low-contrast changes. It proposes WGDF, a network that splits each image with a Discrete Wavelet Transform into a high-frequency branch for local details and a low-frequency branch for global structure, processes the two branches with specialized modules, and fuses them back together. The motivation is that fine-grained differences between two dates are hard to see in ordinary pixels but become visible in frequency components. If the claim is right, applications such as disaster monitoring, urban-expansion tracking, and infrastructure management would get change maps with cleaner boundaries and fewer missed small changes.

What carries the argument

The load-bearing mechanism is the Discrete Wavelet Transform (DWT), which decomposes each input image into a low-frequency approximation and high-frequency detail subbands, creating two parallel feature streams. In the high-frequency stream, the DFFE module strengthens edge representation and the FDID module performs interaction and differencing in the frequency domain to expose subtle changes. In the low-frequency stream, Transformers capture global semantic context and the PCDM progressively refines the coarse change regions. A final fusion recombines the two streams so local edge sensitivity and global structural consistency are both available to the decoder.

What would settle it

Train an otherwise identical network on the same data and budget without the DWT split (or with the split replaced by ordinary spatial gradient filters) and compare F1 or IoU on edge-heavy and subtle-change test subsets; if the single-stream model matches WGDF, the wavelet dual-frequency encoding is not the cause of the reported gains.

Watch

Extended reading notes

Core claim

The paper's central claim is that decomposing each input image with a Discrete Wavelet Transform into high-frequency detail and low-frequency structure improves change detection. The high-frequency branch uses DFFE to strengthen edge detail and FDID to capture fine-grained temporal differences; the low-frequency branch uses Transformers for global context and PCDM to progressively refine change regions. The two streams are fused to combine local sensitivity with global discriminability. WGDF is reported to reduce edge ambiguity and to outperform prior methods on the evaluated remote-sensing datasets.

Load-bearing premise

The load-bearing premise is that wavelet-domain decomposition genuinely adds discriminative information over spatial-domain features, and that the reported benchmark comparisons use identical training and evaluation protocols for all methods.

Editorial extensions

If this is right

  • Edge regions in predicted change maps should become cleaner and more precisely localized because the high-frequency branch is explicitly dedicated to detail.
  • Small or low-contrast changes should be detected more often, since differencing in the frequency domain amplifies fine-grained differences that spatial features suppress.
  • Global consistency should reduce false alarms, because the low-frequency transformer branch keeps long-range context in the loop.
  • The DFFE, FDID, and PCDM modules are modular enough to be dropped into other change-detection backbones, potentially transferring the gains.
  • If the reported benchmark numbers hold, WGDF would set a new state of the art for remote-sensing change detection at the time of publication.

Reading between the lines

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

  • The paper evaluates on standard benchmarks; a natural extension is to test on heterogeneous pairs (for example, optical versus synthetic-aperture radar) to see whether the frequency split survives large domain gaps.
  • The same dual-frequency design could transfer to other pixel-level change tasks, such as video object segmentation or medical follow-up imaging, where boundaries and subtle differences carry the signal.
  • An ablation identifying which wavelet subbands (horizontal, vertical, diagonal) contribute most could let a leaner version keep only the informative bands and reduce compute.
  • The paper does not report a matched-parameter baseline; such a comparison would separate the effect of the frequency decomposition from the effect of added network capacity.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 2 minor

Summary. The paper proposes Wavelet-Guided Dual-Frequency Encoding (WGDF) for remote sensing change detection. The method is described in the abstract as applying a discrete wavelet transform to split input pairs into high-frequency and low-frequency components; a Dual-Frequency Feature Enhancement (DFFE) module and a Frequency-Domain Interactive Difference (FDID) module are used on the high-frequency branch, while a low-frequency branch uses Transformers and a Progressive Contextual Difference Module (PCDM). The two branches are fused to combine local sensitivity with global discriminability. The central claim, stated in the abstract, is that extensive experiments on multiple remote sensing datasets show WGDF significantly alleviates edge ambiguity and achieves superior detection accuracy and robustness compared to state-of-the-art methods. However, the supplied full text is almost entirely corrupted by encoding failures: paragraphs are replaced by replacement glyphs, tables contain no legible numeric entries, and an unrelated arXiv identifier (2508.05239v1 [cs.CL]) appears mid-text. As a result, the method description, experimental setup, quantitative results, ablations, and training protocols cannot be inspected.

Significance. If the proposed architecture and the claimed benchmark results were fully verifiable, the work could be a useful contribution to change detection, as frequency-domain/wavelet-guided modeling is an underexplored direction and the authors commit to releasing code. The architectural combination of DWT high/low-frequency decomposition with a transformer branch is plausible and potentially of interest. That said, the current manuscript provides no readable evidence for any of these contributions: there are no legible equations, no quantifiable experimental tables, no ablations, and no implementation details. The paper does not ship machine-checked proofs, reproducible code, parameter-free derivations, or falsifiable predictions that could be assessed independently. Therefore the significance cannot currently be evaluated beyond the plausibility of the idea.

major comments (3)
  1. [Section 3 (Method)] The descriptions of DFFE, FDID, and PCDM are unreadable in the supplied text: equations and module details are replaced by garbled glyphs. It is impossible to verify the claimed dual-frequency mechanism, the fusion strategy, or the architectural novelty. This is load-bearing because the entire paper is about these modules.
  2. [Section 4 (Experiments, Tables 1-5)] All experimental tables contain no legible numeric values; the cells are filled with corrupted placeholder characters. The central claim of 'superior detection accuracy and robustness compared to state-of-the-art methods' is therefore entirely unsupported by the available text. No F1, IoU, precision/recall, dataset names, data splits, or training protocol can be verified.
  3. [Full manuscript text] The text includes an unrelated arXiv identifier, 'arXiv:2508.05239v1 [cs.CL]', and repeated garbled table captions. This prevents locating the ablation studies, comparison protocols, and implementation details, and means the submission is not internally consistent with the title/abstract. This is not a minor typographical issue but the absence of the evidence needed to evaluate the paper.
minor comments (2)
  1. [Abstract] The phrase 'an amplify fine-grained differences' is a typo; it should be 'can amplify fine-grained differences'.
  2. [Code availability] The GitHub repository link is given as a promise but cannot currently be accessed; please ensure the link is live and contains a working implementation at publication time.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity identified; the central claim is an empirical benchmark comparison and the available text supplies no self-citation chain or fitted-input/prediction identity to reduce.

full rationale

The paper's central claim, as stated in the abstract, is that WGDF 'significantly alleviates edge ambiguity and achieves superior detection accuracy and robustness compared to state-of-the-art methods' on multiple remote sensing datasets. This is a test-set performance comparison against external benchmarks, which is a self-contained empirical evaluation rather than a prediction derived from fitted parameters. The method description, equations, and tables in the supplied text are heavily corrupted by encoding failures, and an unrelated arXiv identifier appears mid-text, so no specific derivation chain can be inspected. However, the circularity rules require quoting the paper and exhibiting a concrete reduction, such as an equation that defines one quantity in terms of another and then reports that quantity as a prediction, or a fitted parameter being renamed as a prediction. No such quotation or reduction is possible from the readable material. There is no visible self-citation that is load-bearing, no imported uniqueness theorem, and no ansatz smuggled in via citation. The absence of readable experimental tables and training protocols is a verifiability problem, not a circularity problem. Accordingly, the honest finding is no significant circularity, with score 0.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

Because the full text is unreadable, this ledger is reconstructed from the abstract. The paper introduces no physical entities, forces, or conserved quantities; DFFE, FDID, and PCDM are named neural modules, i.e., architectural structure rather than postulated entities with falsifiable handles outside the paper. The listed free parameters are design choices that would normally be reported in an implementation-details section, which could not be inspected.

free parameters (3)
  • DWT decomposition level and wavelet basis
    How many levels of the discrete wavelet transform are applied, and which wavelet is used (e.g., Haar, Daubechies), is an implementation choice not stated in the abstract; it determines the high/low frequency split that the whole architecture depends on.
  • Fusion weighting between high- and low-frequency branches
    The abstract says the two branches are 'synergistically fused', but the fusion rule (learned weights, concatenation, addition) and any balancing coefficients are not specified in the abstract.
  • Transformer depth and channel configuration
    The low-frequency branch uses Transformers; depth, heads, and channels are standard free hyperparameters fitted or tuned on data, unstated in the abstract.
assumptions (3)
  • standard math DWT losslessly decomposes an image into high- and low-frequency components that correspond to local details and global structures respectively
    DWT is a standard, invertible transform; the semantic mapping to 'details' versus 'structures' is a modeling assumption.
  • domain assumption Frequency-domain modeling, particularly in the wavelet domain, amplifies fine-grained differences and improves edge-change perception
    This is the motivational hypothesis of the paper (abstract, first paragraph); it is plausible but not proven by the abstract.
  • domain assumption The chosen benchmark datasets and metrics are representative, and the reported SOTA comparisons are fair
    Standard assumption of empirical deep-learning papers; unverifiable from the abstract, and the experiments section could not be read.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Wavelet-Guided Dual-Frequency Encoding for Remote Sensing Change Detection." pith.science (2026). https://pith.science/paper/ETAL34GK

@misc{pith2026250805271,
  author       = {Pith},
  title        = {Pith review of: Wavelet-Guided Dual-Frequency Encoding for Remote Sensing Change Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ETAL34GK}},
  note         = {Machine review of arXiv:2508.05271}
}
read the original abstract

Change detection in remote sensing imagery plays a vital role in various engineering applications, such as natural disaster monitoring, urban expansion tracking, and infrastructure management. Despite the remarkable progress of deep learning in recent years, most existing methods still rely on spatial-domain modeling, where the limited diversity of feature representations hinders the detection of subtle change regions. We observe that frequency-domain feature modeling particularly in the wavelet domain an amplify fine-grained differences in frequency components, enhancing the perception of edge changes that are challenging to capture in the spatial domain. Thus, we propose a method called Wavelet-Guided Dual-Frequency Encoding (WGDF). Specifically, we first apply Discrete Wavelet Transform (DWT) to decompose the input images into high-frequency and low-frequency components, which are used to model local details and global structures, respectively. In the high-frequency branch, we design a Dual-Frequency Feature Enhancement (DFFE) module to strengthen edge detail representation and introduce a Frequency-Domain Interactive Difference (FDID) module to enhance the modeling of fine-grained changes. In the low-frequency branch, we exploit Transformers to capture global semantic relationships and employ a Progressive Contextual Difference Module (PCDM) to progressively refine change regions, enabling precise structural semantic characterization. Finally, the high- and low-frequency features are synergistically fused to unify local sensitivity with global discriminability. Extensive experiments on multiple remote sensing datasets demonstrate that WGDF significantly alleviates edge ambiguity and achieves superior detection accuracy and robustness compared to state-of-the-art methods. The code will be available at https://github.com/boshizhang123/WGDF.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 52 canonical work pages

  1. [1]

    G. Xu, Y. Liu, L. Deng, X. Wang, and H. Zhu, ``Smnet: A semantic guided mamba network for remote sensing change detection,'' IEEE Transactions on Aerospace and Electronic Systems, 2025

    work page 2025

  2. [2]

    J. Lu, Q. Hu, R. Zhu, and H. Jia, ``A high-resolution remote sensing image registration method combining object and point features,'' IEEE Transactions on Aerospace and Electronic Systems, vol. 58, no. 5, pp. 4196--4213, 2022

    work page 2022

  3. [3]

    C. Wu, B. Du, and L. Zhang, ``Fully convolutional change detection framework with generative adversarial network for unsupervised, weakly supervised and regional supervised change detection,'' IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45, no. 8, pp. 9774--9788, 2023

    work page 2023

  4. [4]

    K. Li, Z. Liu, and Z. Zhang, ``Region-based global-local contrastive representation learning for fine-grained ship classification in remote sensing images,'' IEEE Transactions on Aerospace and Electronic Systems, 2024

    work page 2024

  5. [5]

    Yang, K.-M

    Y. Yang, K.-M. Lam, J. Dong, and Y. Ju, ``Multi-factor deep learning model for sea surface temperature forecasting,'' Remote Sensing, vol. 17, no. 5, p. 752, 2025

    work page 2025

  6. [6]

    L. Miao, X. Li, X. Zhou, L. Yao, Y. Deng, T. Hang, Y. Zhou, and H. Yang, ``Snunet3+: A full-scale connected siamese network and a dataset for cultivated land change detection in high-resolution remote-sensing images,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 62, 2023

    work page 2023

  7. [7]

    Bovolo and L

    F. Bovolo and L. Bruzzone, ``A theoretical framework for unsupervised change detection based on change vector analysis in the polar domain,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 45, no. 1, pp. 218--236, 2006

    work page 2006

  8. [8]

    C. Yang, J. H. Everitt, and J. M. Bradford, ``Yield estimation from hyperspectral imagery using spectral angle mapper (sam),'' Transactions of the ASABE, vol. 51, no. 2, pp. 729--737, 2008

    work page 2008

Show all 53 references
  1. [9]

    Celik, ``Unsupervised change detection in satellite images using principal component analysis and k -means clustering,'' IEEE geoscience and remote sensing letters, vol

    T. Celik, ``Unsupervised change detection in satellite images using principal component analysis and k -means clustering,'' IEEE geoscience and remote sensing letters, vol. 6, no. 4, pp. 772--776, 2009

    2009

  2. [10]

    Wang and C.-I

    J. Wang and C.-I. Chang, ``Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis,'' IEEE transactions on geoscience and remote sensing, vol. 44, no. 6, pp. 1586--1600, 2006

    2006

  3. [11]

    T. V. Bandos, L. Bruzzone, and G. Camps-Valls, ``Classification of hyperspectral images with regularized linear discriminant analysis,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 47, no. 3, pp. 862--873, 2009

    2009

  4. [12]

    Zhang, X

    K. Zhang, X. Zhao, F. Zhang, L. Ding, J. Sun, and L. Bruzzone, ``Relation changes matter: Cross-temporal difference transformer for change detection in remote sensing images,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1--15, 2023

    2023

  5. [13]

    R. C. Daudt, B. Le Saux, and A. Boulch, ``Fully convolutional siamese networks for change detection,'' in 2018 25th IEEE international conference on image processing (ICIP). 1em plus 0.5em minus 0.4em IEEE, 2018, pp. 4063--4067

    2018

  6. [14]

    S. Zhao, X. Zhang, P. Xiao, and G. He, ``Exchanging dual-encoder--decoder: A new strategy for change detection with semantic guidance and spatial localization,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1--16, 2023

    2023

  7. [15]

    Y. Liu, F. Zhang, S. Zhang, K. Zhang, J. Sun, and L. Bruzzone, ``Content-guided spatial-spectral integration network for change detection in hr remote sensing images,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 62, 2024

    2024

  8. [16]

    Jiang, J

    K. Jiang, J. Liu, W. Zhang, F. Liu, and L. Xiao, ``Manet: An efficient multi-dimensional attention-aggregated network for remote sensing image change detection,'' IEEE Transactions on Geoscience and Remote Sensing, 2023

    2023

  9. [17]

    H. Chen, Z. Qi, and Z. Shi, ``Remote sensing image change detection with transformers,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1--14, 2021

    2021

  10. [18]

    H. Guo, X. Su, C. Wu, B. Du, and L. Zhang, ``Saan: Similarity-aware attention flow network for change detection with vhr remote sensing images,'' IEEE Transactions on Image Processing, 2024

    2024

  11. [19]

    Z. Li, S. Cao, J. Deng, F. Wu, R. Wang, J. Luo, and Z. Peng, ``Stade-cdnet: Spatial--temporal attention with difference enhancement-based network for remote sensing image change detection,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1--17, 2024

    2024

  12. [20]

    Huang, Q

    J. Huang, Q. Shen, M. Wang, and M. Yang, ``Multiple attention siamese network for high-resolution image change detection,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1--16, 2021

    2021

  13. [21]

    X. Xu, Z. Yang, and J. Li, ``Amca: Attention-guided multiscale context aggregation network for remote sensing image change detection,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1--19, 2023

    2023

  14. [22]

    P. Li, T. Si, C. Ye, and Q. Guo, ``Semantic-aware transformer with feature integration for remote sensing change detection,'' Engineering Applications of Artificial Intelligence, vol. 135, p. 108774, 2024

    2024

  15. [23]

    Jiang, Z

    B. Jiang, Z. Wang, X. Wang, Z. Zhang, L. Chen, X. Wang, and B. Luo, ``Vct: Visual change transformer for remote sensing image change detection,'' IEEE Transactions on Geoscience and Remote Sensing, 2023

    2023

  16. [24]

    X. Li, Y. Tan, K. Liu, X. Wang, and X. Zhou, ``Dsfi-cd: Diffusion-driven space-frequency domain information interaction for remote sensing image change detection,'' IEEE Transactions on Geoscience and Remote Sensing, 2025

    2025

  17. [25]

    Y. Liu, Y. Ju, Y. Rao, H. Fan, J. Dong, F. Gao, and Q. Du, ``Aerial multi-view stereo via adaptive depth range inference and normal cues,'' IEEE Transactions on Geoscience and Remote Sensing, 2025

    2025

  18. [26]

    Y. Zhan, K. Fu, M. Yan, X. Sun, H. Wang, and X. Qiu, ``Change detection based on deep siamese convolutional network for optical aerial images,'' IEEE Geoscience and Remote Sensing Letters, vol. 14, no. 10, pp. 1845--1849, 2017

    2017

  19. [27]

    Y. Liu, C. Pang, Z. Zhan, X. Zhang, and X. Yang, ``Building change detection for remote sensing images using a dual-task constrained deep siamese convolutional network model,'' IEEE Geoscience and Remote Sensing Letters, vol. 18, no. 5, pp. 811--815, 2020

    2020

  20. [28]

    Chen and Z

    H. Chen and Z. Shi, ``A spatial-temporal attention-based method and a new dataset for remote sensing image change detection,'' Remote Sensing, vol. 12, no. 10, p. 1662, 2020

    2020

  21. [29]

    S. Fang, K. Li, J. Shao, and Z. Li, ``Snunet-cd: A densely connected siamese network for change detection of vhr images,'' IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1--5, 2021

    2021

  22. [30]

    W. G. C. Bandara and V. M. Patel, ``A transformer-based siamese network for change detection,'' in IGARSS 2022-2022 IEEE International Geoscience and Remote Sensing Symposium. 1em plus 0.5em minus 0.4em IEEE, 2022, pp. 207--210

    2022

  23. [31]

    Y. Feng, H. Xu, J. Jiang, H. Liu, and J. Zheng, ``Icif-net: Intra-scale cross-interaction and inter-scale feature fusion network for bitemporal remote sensing images change detection,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1--13, 2022

    2022

  24. [32]

    Zheng, A

    Z. Zheng, A. Ma, L. Zhang, and Y. Zhong, ``Change is everywhere: Single-temporal supervised object change detection in remote sensing imagery,'' in Proceedings of the IEEE/CVF international conference on computer vision, 2021, pp. 15\,193--15\,202

    2021

  25. [33]

    Y. Feng, J. Jiang, H. Xu, and J. Zheng, ``Change detection on remote sensing images using dual-branch multilevel intertemporal network,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1--15, 2023

    2023

  26. [34]

    Noman, M

    M. Noman, M. Fiaz, H. Cholakkal, S. Khan, and F. S. Khan, ``Elgc-net: Efficient local-global context aggregation for remote sensing change detection,'' IEEE Transactions on Geoscience and Remote Sensing, 2024

    2024

  27. [35]

    J. Leng, Y. Ju, Y. Duan, J. Zhang, Q. Lv, Z. Wu, and H. Fan, ``Fnin: A fourier neural operator-based numerical integration network for surface-from-gradients,'' in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 39, no. 5, 2025, pp. 4580--4588

    2025

  28. [36]

    Y. Ju, J. Xiao, C. Zhang, H. Xie, A. Luo, H. Zhou, J. Dong, and A. C. Kot, ``Towards marine snow removal with fusing fourier information,'' Information Fusion, vol. 117, p. 102810, 2025

    2025

  29. [37]

    Y. Ju, B. Shi, B. Wen, K.-M. Lam, X. Jiang, and A. C. Kot, ``Revisiting one-stage deep uncalibrated photometric stereo via fourier embedding,'' IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025

    2025

  30. [38]

    Zheng, M

    H. Zheng, M. Gong, T. Liu, F. Jiang, T. Zhan, D. Lu, and M. Zhang, ``Hfa-net: High frequency attention siamese network for building change detection in vhr remote sensing images,'' Pattern Recognition, vol. 129, p. 108717, 2022

    2022

  31. [39]

    Cheng, H

    H. Cheng, H. Wu, J. Zheng, K. Qi, and W. Liu, ``A hierarchical self-attention augmented laplacian pyramid expanding network for change detection in high-resolution remote sensing images,'' ISPRS Journal of Photogrammetry and Remote Sensing, vol. 182, pp. 52--66, 2021

    2021

  32. [40]

    Liang, Z

    S. Liang, Z. Hua, and J. Li, ``Enhanced self-attention network for remote sensing building change detection,'' IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 16, pp. 4900--4915, 2023

    2023

  33. [41]

    Y. Zhou, C. Huo, J. Zhu, L. Huo, and C. Pan, ``Dcat: Dual cross-attention-based transformer for change detection,'' Remote Sensing, vol. 15, no. 9, p. 2395, 2023

    2023

  34. [42]

    F. Gao, X. Wang, J. Dong, and S. Wang, ``Synthetic aperture radar image change detection based on frequency-domain analysis and random multigraphs,'' Journal of Applied Remote Sensing, vol. 12, no. 1, pp. 016\,010--016\,010, 2018

    2018

  35. [43]

    C. Zhao, L. Ma, L. Wang, T. Ohtsuki, P. T. Mathiopoulos, and Y. Wang, ``Sar image change detection in spatial-frequency domain based on attention mechanism and gated linear unit,'' IEEE Geoscience and Remote Sensing Letters, vol. 20, pp. 1--5, 2023

    2023

  36. [44]

    M. Cai, H. Zhang, H. Huang, Q. Geng, Y. Li, and G. Huang, ``Frequency domain image translation: More photo-realistic, better identity-preserving,'' in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 13\,930--13\,940

    2021

  37. [45]

    Y. Liu, R. Yu, J. Wang, X. Zhao, Y. Wang, Y. Tang, and Y. Yang, ``Global spectral filter memory network for video object segmentation,'' in European Conference on Computer Vision. 1em plus 0.5em minus 0.4em Springer, 2022, pp. 648--665

    2022

  38. [46]

    Y. Yang, L. Jiao, X. Liu, F. Liu, S. Yang, L. Li, P. Chen, X. Li, and Z. Huang, ``Dual wavelet attention networks for image classification,'' IEEE Transactions on Circuits and Systems for Video Technology, vol. 33, no. 4, pp. 1899--1910, 2022

    1910

  39. [47]

    Y. Tang, S. Feng, C. Zhao, Y. Fan, Q. Shi, W. Li, and R. Tao, ``An object fine-grained change detection method based on frequency decoupling interaction for high-resolution remote sensing images,'' IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1--13, 2023

    2023

  40. [48]

    T. Yan, Z. Wan, and P. Zhang, ``Fully transformer network for change detection of remote sensing images,'' in Proceedings of the Asian Conference on Computer Vision, 2022, pp. 1691--1708

    2022

  41. [49]

    D. P. Kingma, ``Adam: A method for stochastic optimization,'' arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  42. [50]

    Zhang, Z

    J. Zhang, Z. Shao, Q. Ding, X. Huang, Y. Wang, X. Zhou, and D. Li, ``Aernet: An attention-guided edge refinement network and a dataset for remote sensing building change detection,'' IEEE Transactions on Geoscience and Remote Sensing, 2023

    2023

  43. [51]

    Huang, X

    Y. Huang, X. Li, Z. Du, and H. Shen, ``Spatiotemporal enhancement and interlevel fusion network for remote sensing images change detection,'' IEEE Transactions on Geoscience and Remote Sensing, 2024

    2024

  44. [52]

    W. Wang, C. Liu, G. Liu, and X. Wang, ``Cf-gcn: Graph convolutional network for change detection in remote sensing images,'' IEEE Transactions on Geoscience and Remote Sensing, 2024

    2024

  45. [53]

    Noman, M

    M. Noman, M. Fiaz, and H. Cholakkal, ``Changebind: A hybrid change encoder for remote sensing change detection,'' in IGARSS 2024-2024 IEEE International Geoscience and Remote Sensing Symposium. 1em plus 0.5em minus 0.4em IEEE, 2024, pp. 8417--8422

    2024

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.