Pith. sign in

REVIEW 3 major objections 5 minor 45 references

Warping Earth Observations for better ice labeling in the Marginal Marginal Ice Zone

T0 review · 3 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read Aligning satellite images before fusion lifts sea ice accuracy to 0.88

desk verdict Useful sparse-label Antarctic MIZ dataset and a plausible LMI-warping claim that needs error bars, dense metrics, and a control warp before the causal story holds. read the letter →

arxiv 2608.11883 v1 pith:YONMLMXH submitted 2026-08-12 cs.CV

Tom Kelly , Martin S. J. Rogers This is my paper · ORCID
classification cs.CV
keywords seaicesegmentationmultimodalregistrationmutualinformationEarthobservationSentinel-1MODISimagewarpingsparselabels
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

This paper argues that multimodal satellite images of fast-moving scenes should be geometrically aligned before any fusion or classification, then shows that doing so with a learned warp field substantially improves ice-water segmentation in the Antarctic marginal ice zone. The authors introduce a sparse expert-labeled dataset of 2,088 pixel pins and demonstrate that warping MODIS imagery onto Sentinel-1 radar coordinates raises balanced accuracy from about 0.86 to 0.88, close to the 0.91 ceiling set by human expert agreement. If the claim holds, pre-fusion spatial grounding becomes a generally useful step for any Earth-observation pipeline that combines sensors observing a moving surface.

What carries the argument

The load-bearing component is a composite loss whose main term is Local Mutual Information (LMI), computed over 31×31 pixel neighborhoods between the radar and optical channels, which drives a U-Net to predict a dense per-pixel displacement field. The field is applied through a spatial transformer network to warp all 38 MODIS channels up to Sentinel-1 resolution and coordinates. Additional physics-informed regularizers—Jacobian area preservation, smoothness, a displacement cap, and a zero-movement prior—keep the warp consistent with real sea ice drift, while a hyperparameter sweep over 152 configurations selects the weights and the channel pair (HH versus MODIS band 2) that gives the cleanest class separation.

What would settle it

Take a held-out test scene with known ice drift from GPS buoys or feature tracking, run the learned warp, and compare the predicted displacement field to the measured drift; if the warp field diverges from measured drift while still improving classification, the gains come from feature transformation rather than true alignment.

Watch

Extended reading notes

Core claim

The central discovery is that a deformable warp field learned by maximizing local mutual information between Sentinel-1 HH backscatter and MODIS band 2, then applied to all MODIS channels, realigns ice-water features across modalities enough to improve downstream pixel-wise and deep-learning classifiers. Across the full 56-channel set, most models improve with warping, with the best reaching 0.88 balanced accuracy versus a 0.91 oracle baseline, and the gains are largest at the difficult ice-water boundary. The paper reads this as evidence that explicit perceptual grounding—modeling where each sensor's observation actually points—adds complementary information that multimodal fusion alone does not provide.

Load-bearing premise

The entire improvement rests on the assumption that maximizing local mutual information between one radar channel and one optical channel signals true geometric correspondence of ice and water features, rather than some radiometric coincidence.

Editorial extensions

If this is right

  • Multimodal Earth-observation pipelines for dynamic scenes should treat pixel correspondence as something to be learned, not assumed.
  • Sparse, high-quality expert labels at boundaries can train dense segmentation models when the input modalities are aligned first.
  • Geospatial foundation models that assume static co-registration will increasingly misalign as spatial resolution grows, making motion-aware grounding a needed pretraining component.
  • The warp field itself could double as a dense sea-ice drift estimate, useful for forecasting and motion studies.

Reading between the lines

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

  • The method's dependence on a single channel pair for the warp suggests a testable boundary: scenes where ice appears differently in HH and band 2 (e.g., melt ponds, new ice) may produce unreliable warps; using multi-channel LMI could be more robust.
  • Because the regularizers encode physical drift limits, the same architecture could transfer to other dynamic surfaces like flood water or wildfire smoke with adjusted physical priors.
  • A natural extension is to jointly learn the warp and the segmentation in an end-to-end fashion; the paper's preprocessing separation may under-use the alignment signal.
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 / 5 minor

Summary. The paper proposes an unsupervised mutual-information-based warping network that aligns lower-resolution MODIS imagery to higher-resolution Sentinel-1 SAR imagery before multimodal fusion, with the goal of improving pixel-wise sea ice/water segmentation in the Antarctic marginal ice zone. The authors introduce a sparse expert-labeled dataset of 2,088 pins across 43 scenes, evaluate several single-pixel classifiers and two UNet variants with and without warping, and report that warping improves balanced accuracy for most models, achieving a maximum bAcc of 0.88 versus an oracle of 0.91. The central claim is that explicitly correcting spatial misalignment across modalities prior to fusion improves downstream segmentation and that sparse, point-wise labels can train dense segmentation models.

Significance. If the central claim is established, the work addresses a real and underappreciated problem in multimodal Earth observation: the pixel-level correspondence assumption is violated in dynamic scenes such as the sea ice marginal zone. The introduced sparse expert-labeled dataset is a useful resource, and the idea of using local mutual information for cross-modal, cross-resolution deformable registration is interesting. The paper also ships a public code repository (footnote 1), which supports reproducibility. However, the strength of the empirical claim is currently limited by the absence of statistical uncertainty quantification and by the lack of a control warp that would isolate geometric alignment as the causal mechanism.

major comments (3)
  1. [Section 6, Tables 3 and 4] The central conclusion that warping 'significantly improved' segmentation (Section 6) is not supported by any statistical inference. Tables 3 and 4 report point estimates only, with no confidence intervals, per-scene variability, or significance tests. The reported gains are small for the best models (e.g., UNet B: 0.86106 to 0.88364 in Table 4) and one model (RBF SVM in Table 3) degrades after warping (0.7907 to 0.7873). Because the test set is small (43 scenes split 60/20/20, so roughly 8-9 test scenes), the differences could plausibly be noise. The paper should provide confidence intervals or significance tests (e.g., bootstrapping over scenes or pins) before claiming improvement.
  2. [Section 3.1, Equations (1)-(6) and Section 6] The causal claim that geometric alignment is the reason for improved segmentation is not established, because the only comparison is warped versus unwarped features. The warp is a high-capacity UNet optimized to maximize local mutual information between exactly one channel pair (S1:HH and MODIS:2), and the warp hyperparameters and channel pair were selected using validation labels (Section 4.6, Table 5). The observed improvements could be an artifact of a task-specific feature transform that happens to move MODIS edge statistics toward the labeled boundary region, rather than a correction of physical misalignment. To support the alignment interpretation, the paper needs a control warp with comparable flexibility that does not correspond to physical correspondence (e.g., a random smooth warp, a warp trained on a non-corresponding channel pair, or an inverted warp) and should show that the estimated displacement fields are consistent with physical ice drift (e.g., by comparing to independently estimated drift or by reporting displacement accuracy, not just the validation margin in Table 5).
  3. [Section 4.6 and Section 4.4] The hyperparameter selection procedure raises a risk of overfitting to the validation partition, and the paper does not quantify this risk. The 152-configuration Bayesian sweep in Section 4.6 selects warp hyperparameters and the LMI channel pair using the validation mean signed distance (an LSVM margin), and the UNet architectures A and B in Section 4.4 are also selected on the validation partition. Since the reported test results are point estimates without error bars, it is unclear how much of the observed improvement is due to the method itself rather than to selection on the same validation labels. The paper should either report the distribution of validation/test performance across the sweep, or use a nested validation procedure, and should at minimum report the number of test pins and per-scene results.
minor comments (5)
  1. [Table 3] The 'All Ice Baseline' row reports bAcc=0.5000 when unwarped and bAcc=0.2009 when warped. Since an all-ice prediction is independent of the input features, this difference is unexpected and may indicate a typo or an evaluation inconsistency; please clarify.
  2. [Section 3.2] The paper states that 56 channels are used in the multivariate experiments, but the described feature set (2 S1, 38 MODIS, 16 AMSR, 2 topography) sums to 58. Please reconcile this number.
  3. [Title] The title contains a duplicated word: 'Marginal Marginal Ice Zone'. This is likely a typo.
  4. [Section 2] Reference [32] is cited as evidence that SAR backscatter is difficult to interpret for sea ice analysis, but the reference concerns incidence angle normalization over agricultural surfaces. A more appropriate sea-ice-specific reference would strengthen the argument.
  5. [Figures 5 and 6] The dense labeling visualizations are qualitative. Since the paper claims dense segmentation from sparse labels, a quantitative evaluation of dense output (e.g., using any available ice charts or a separate dense test set) would be more convincing than visual inspection alone.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the warp is unsupervised, the test partition is held out, and validation sweeps are standard model selection.

full rationale

The claimed derivation chain is: (1) an unsupervised per-scene UNet warp is trained by maximizing local mutual information between S1:HH and MODIS:2 plus regularization terms (Eqs. 1-6); (2) the warp is applied to MODIS channels to form aligned features; (3) pixel-wise and UNet classifiers are trained on warped versus unwarped features and evaluated on a held-out test partition. The warp objective contains no label term, so the displacement field is not constructed from the classification target. The warp hyperparameters and the S1:HH-MODIS:2 channel pair are selected by a validation-set LSVM margin (Section 4.6, Table 5); this is model selection, not a fitted prediction, and the reported bAcc gains (Tables 3-4) are computed on the independent test split described in Section 3.2. The only overlapping-author citation is Rogers et al. [35], used to support the ancillary claim that SAR-plus-visible fusion improves separability; it is not load-bearing for the warping result. The absence of a random or mismatched warp control weakens the causal interpretation that geometric alignment rather than feature transformation drives the uplift, but that is an experimental confound, not circularity: no equation or fitted parameter is equivalent to the reported improvement by construction.

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

The central claim depends on domain assumptions about sea ice physics and cross-modal similarity (incompressibility, smoothness, bounded drift, and the validity of LMI between S1:HH and MODIS band 2), plus hyperparameters tuned on the validation split. No new physical entities are introduced.

free parameters (4)
  • lambda_jac, lambda_smooth, lambda_cap, lambda_zero = 1e-3, 5.72e-2, 9.872, 6.946e-7
    Chosen by Bayesian sweep of 152 parameter sets to maximize validation-class separation (Section 4.6).
  • u_max (displacement cap threshold) = not reported
    Introduced in Eq. 5 as 'predefined pixel threshold'; value not stated in the paper, yet caps ice drift.
  • LMI binning parameters (soft Gaussian histogram width/count) = not reported
    Used in Eq. 2 to compute Shannon entropy over 31x31 windows; values are not specified, affecting the warp field.
  • UNet architecture and training hyperparameters = model A/B selected by sweep
    Best architectures for warped/unwarped data selected from 256 experiments on the validation split (Section 4.4), affecting final reported bAcc.
assumptions (5)
  • domain assumption Sea ice drift between the S1 and MODIS acquisitions is approximately incompressible, smooth, bounded, and near-zero in ambiguous regions.
    Enforced through L_jac, L_smooth, L_cap, L_zero in Eq. 1; if the real deformation violates these, the warp field will be physically wrong.
  • domain assumption Local mutual information between S1:HH backscatter and MODIS channel 2 over a 31x31 window is a reliable similarity signal for cross-modal alignment.
    This is the only image-similarity signal used to train the warp (Eq. 2); no validation against ground-truth drift is performed.
  • domain assumption The 2,088 sparse expert pins sampled near the ice-water interface are representative ground truth for point classification and dense segmentation quality.
    Pins were placed by sampling MODIS channel 1 around the boundary (Section 3.2), and dense segmentation is only qualitatively assessed.
  • domain assumption A one-hour maximum time gap between acquisitions bounds ice drift enough that a single warp field can recover alignment.
    Used in Section 3.2 to justify the setup; if drift exceeds the warp field's capacity, alignment fails.
  • standard math Train/validation/test splits are made by geographic chip so that pins in the same scene do not leak across partitions.
    Standard supervised-learning assumption; stated in Section 3.2 but not empirically verified.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Warping Earth Observations for better ice labeling in the Marginal Marginal Ice Zone." pith.science (2026). https://pith.science/paper/YONMLMXH

@misc{pith2026260811883,
  author       = {Pith},
  title        = {Pith review of: Warping Earth Observations for better ice labeling in the Marginal Marginal Ice Zone},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YONMLMXH}},
  note         = {Machine review of arXiv:2608.11883}
}
read the original abstract

Multimodal satellite imagery provides complementary information for Earth Observation, but accurately combining heterogeneous sensors remains challenging in dynamic environments. Fast-changing regions, such as the Antarctic marginal ice zone, cannot fully exploit multimodal information from different satellite sensors because surface features move between image acquisitions. This spatial and temporal mismatch challenges effective perceptual grounding, violating the assumption of pixel-level correspondence that underpins most multimodal reasoning and downstream classification pipelines. Antarctic sea ice provides a challenging benchmark due to the rapid, heterogeneous drift of individual ice floes and the differing responses of sea ice to radar, visible and thermal sensing modalities. Accurate, dense supervision of sea ice remains scarce because generating pixel-wise labels requires time-consuming expert interpretation of noisy data, leading to historical reliance on coarse-resolution maritime ice charts for model training. This paper presents a novel architecture based on mutual information warping to align multi-satellite (Sentinel-1 and MODIS platforms) multimodal (visible, thermal, radar) satellite scenes. To demonstrate the approach, we introduce a sparse expert-labeled dataset of 2,088 pixel-wise annotations (7,046 expert point classifications) located at the ice-water margin interface across 43 scenes. Our results demonstrate that spatially grounding and aligning modalities prior to segmentation improves classification accuracy, and enables accurate, dense sea ice segmentation from sparse point-wise supervision.

Figures

Figures reproduced from arXiv: 2608.11883 by the authors.

Figure 1
Figure 1. Different satellite sources image Antarctica at various locations and scales, taken at similar, but not identical, times (a, b), leading to a lack of alignment on fast￾moving objects such as sea ice across modalities (c). We demonstrate that by aligning these modalities with warping, we improve learning results; (d) shows the ice masks from one satellite (MODIS, blue) and a second (S1, gray) before our warping techn… view at source ↗
Figure 2
Figure 2. In the first stage (solid arrows), we warp the MODIS channels to match the S1 channels using a warp field created by a UNet. This field is applied to the MODIS channels to create an aligned MODIS image. This optimization takes place on a loss, L. In the second stage the warped dataset is built from the pixel feature vectors (xn) with the expert’s ice/water labels (circles, y). We observe that ice is more easily iden… view at source ↗
Figure 3
Figure 3. The user interface (above) and the MODIS channel 1 used to distribute pins over the marginal ice (below: left shows same scene as above). The users can zoom and pan with the mouse. This dataset results in binary labels with point-wise classifications and corre￾sponding satellite features per point. While not all experts labeled every pin, we observed strong agreement ( [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Dataset statistics demonstrating strong agreement among experts (left) and non-trivial overlap between classifications and two representative modalities (right). 4 Experiments [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Input RGB shows S1:HH, S1:HV, and MODIS channels. The label results show S1:HH in gray and the segmentation in cyan. M:1 LSVM is the first MODIS channel only. The other LSVM and GB columns are created with all channels [PITH_FULL_IMAGE:figures/full_fig_p013_5.png]
Figure 6
Figure 6. Figure 6: As [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 38 canonical work pages

  1. [1]

    Marine Policy75, 300–317 (2017)

    Aksenov, Y., Popova, E.E., Yool, A., Nurser, A.G., Williams, T.D., Bertino, L., Bergh, J.: On the future navigability of arctic sea routes: High-resolution projec- tions of the arctic ocean and sea ice. Marine Policy75, 300–317 (2017)

    work page 2017

  2. [2]

    Journal of Geophysical Research: Oceans125(3), e2019JC015418 (2020)

    Alberello, A., Bennetts, L., Heil, P., Eayrs, C., Vichi, M., MacHutchon, K., Ono- rato, M., Toffoli, A.: Drift of pancake ice floes in the winter antarctic marginal ice zone during polar cyclones. Journal of Geophysical Research: Oceans125(3), e2019JC015418 (2020)

    work page 2020

  3. [3]

    In: Proceedings of the 1st ACM SIGSPATIAL International Workshop on Polar Data Science

    Alkaee Taleghan, S., Karimzadeh, M., Barrett, A.P., Meier, W.N., Banaei-Kashani, F.: Ice-fmbench a foundation model benchmark for sea ice type segmentation. In: Proceedings of the 1st ACM SIGSPATIAL International Workshop on Polar Data Science. pp. 1–10 (2025)

    work page 2025

  4. [4]

    IEEE transactions on medical imaging38(8), 1788–1800 (2019)

    Balakrishnan, G., Zhao, A., Sabuncu, M.R., Guttag, J., Dalca, A.V.: Voxelmorph: a learning framework for deformable medical image registration. IEEE transactions on medical imaging38(8), 1788–1800 (2019)

    2019

  5. [5]

    Remote Sensing12(13), 2165 (2020)

    Boulze, H., Korosov, A., Brajard, J.: Classification of sea ice types in sentinel-1 sar data using convolutional neural networks. Remote Sensing12(13), 2165 (2020)

    work page 2020

  6. [6]

    Medical image analysis77, 102339 (2022)

    Chen, J., Frey, E.C., He, Y., Segars, W.P., Li, Y., Du, Y.: Transmorph: Transformer for unsupervised medical image registration. Medical image analysis77, 102339 (2022)

    work page 2022

  7. [7]

    The Cryosphere18(4), 1621–1632 (2024)

    Chen,X.,Patel,M.,PenaCantu,F.J.,Park,J.,NoaTurnes,J.,Xu,L.,Scott,K.A., Clausi, D.A.: Mmseaice: a collection of techniques for improving sea ice mapping with a multi-task model. The Cryosphere18(4), 1621–1632 (2024)

    work page 2024

  8. [8]

    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing14, 5831–5841 (2021)

    De Gelis, I., Colin, A., Longépé, N.: Prediction of categorized sea ice concentration from sentinel-1 sar images based on a fully convolutional network. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing14, 5831–5841 (2021)

    work page 2021

Show all 45 references
  1. [9]

    Medical & Biological Engineering & Computing62(2), 505–519 (2024)

    Deng, L., Lan, Q., Zhi, Q., Huang, S., Wang, J., Yang, X.: Deep learning-based 3d brain multimodal medical image registration. Medical & Biological Engineering & Computing62(2), 505–519 (2024)

    2024

  2. [10]

    Remote Sensing12(9), 1402 (2020)

    Farooq, U., Rack, W., McDonald, A., Howell, S.: Long-term analysis of sea ice drift in the western ross sea, antarctica, at high and low spatial resolution. Remote Sensing12(9), 1402 (2020)

    2020

  3. [11]

    Journal of biomedical optics29(6), 066007–066007 (2024)

    Feenstra, L., Lambregts, M., Ruers, T.J., Dashtbozorg, B.: Deformable multi- modal image registration for the correlation between optical measurements and histology images. Journal of biomedical optics29(6), 066007–066007 (2024)

    2024

  4. [12]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Feng, Z., Atzberger, C., Jaffer, S., Knezevic, J., Sormunen, S., Young, R., Lisaius, M.C., Immitzer, M., Jackson, T., Ball, J., et al.: Tessera: Temporal embeddings of surface spectra for earth representation and analysis. In: Proceedings of the IEEE/CVF Conference on Computer...

    2026

  5. [13]

    ISPRS Journal of Photogrammetry and Remote Sensing230, 258–274 (2025)

    Gao, T., Lan, C., Zhou, C., Zhang, Y., Huang, W., Wang, Y., Wang, L.: Arctic sea ice motion retrieval from multisource sar images using a keypoint-free feature tracking algorithm. ISPRS Journal of Photogrammetry and Remote Sensing230, 258–274 (2025)

    2025

  6. [14]

    Remote sens- ing of Environment202, 18–27 (2017) 16 T

    Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., Moore, R.: Google earth engine: Planetary-scale geospatial analysis for everyone. Remote sens- ing of Environment202, 18–27 (2017) 16 T. Kelly and M. S. J. Rogers

    2017

  7. [15]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Guo, X., Lao, J., Dang, B., Zhang, Y., Yu, L., Ru, L., Zhong, L., Huang, Z., Wu, K., Hu, D., et al.: Skysense: A multi-modal remote sensing foundation model towards universal interpretation for earth observation imagery. In: Proceedings of the IEEE/CVF Conference on Computer V...

    2024

  8. [16]

    arXiv preprint arXiv:2509.13629 (2025)

    He, Y., Liu, M., Liu, Q., Wang, J., Wang, Y., Zhang, H., Chen, X.: Samir, an effi- cient registration framework via robust feature learning from sam. arXiv preprint arXiv:2509.13629 (2025)

    2025

  9. [17]

    ISPRS Journal of Photogrammetry and Remote Sensing238, 624–649 (2026)

    Heffring, M., Xu, L.L.: Geographically-weighted weakly supervised bayesian high- resolution transformer for 200 m resolution pan-arctic sea ice concentration map- ping and uncertainty estimation using sentinel-1, rcm, and amsr2 data. ISPRS Journal of Photogrammetry and Remote ...

    2026

  10. [18]

    The Cryosphere16(3), 1125–1139 (2022)

    Howell, S.E., Brady, M., Komarov, A.S.: Generating large-scale sea ice motion from sentinel-1 and the radarsat constellation mission using the environment and climate change canada automated sea ice tracking system. The Cryosphere16(3), 1125–1139 (2022)

    2022

  11. [19]

    Advances in neural information processing systems28(2015)

    Jaderberg, M., Simonyan, K., Zisserman, A., et al.: Spatial transformer networks. Advances in neural information processing systems28(2015)

    2015

  12. [20]

    arXiv preprint arXiv:2603.01576 (2026)

    Kaushik, S., Maurya, L., Tellman, B., Marsocci, V.: Cryo-bench: Benchmarking foundation models for cryosphere applications. arXiv preprint arXiv:2603.01576 (2026)

    2026 arXiv

  13. [21]

    Remote Sensing Letters17(9), 1027–1040 (2026)

    Khachatrian, E., Lohse, J., Karlsen, T., Dierking, W., Johansson, M.: When classes look alike: leveraging multifrequency sar and tir for separating early-stage sea ice types. Remote Sensing Letters17(9), 1027–1040 (2026)

    2026

  14. [22]

    arXiv preprint arXiv:1412.6980 (2014)

    Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  15. [23]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Klemmer, K., Rolf, E., Robinson, C., Mackey, L., Rußwurm, M.: Satclip: Global, general-purpose location embeddings with satellite imagery. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 39, pp. 4347–4355 (2025)

    2025

  16. [24]

    König, C., König, T., Singha, S., Frost, A., Jacobsen, S.: Combined use of space borneopticalandsardatatoimproveknowledgeaboutseaiceforshipping.Remote Sensing13(23), 4842 (2021)

    2021

  17. [25]

    Remote Sensing9(3), 258 (2017)

    Korosov, A.A., Rampal, P.: A combination of feature tracking and pattern match- ing with optimal parametrization for sea ice drift retrieval from sar data. Remote Sensing9(3), 258 (2017)

    2017

  18. [26]

    Medical physics50(2), 970–979 (2023)

    Lee, D., Alam, S., Jiang, J., Cervino, L., Hu, Y.C., Zhang, P.: Seq2morph: A deep learning deformable image registration algorithm for longitudinal imaging studies and adaptive radiotherapy. Medical physics50(2), 970–979 (2023)

    2023

  19. [27]

    EGUsphere2024, 1–36 (2024)

    Li, W., Hsu, C.Y., Tedesco, M.: Advancing arctic sea ice remote sensing with ai and deep learning: now and future. EGUsphere2024, 1–36 (2024)

    2024

  20. [28]

    International Journal of Remote Sensing44(17), 5344–5374 (2023)

    Pires de Lima, R., Vahedi, B., Hughes, N., Barrett, A.P., Meier, W., Karimzadeh, M.: Enhancing sea ice segmentation in sentinel-1 images with atrous convolutions. International Journal of Remote Sensing44(17), 5344–5374 (2023)

    2023

  21. [29]

    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing17, 13162– 13168 (2024)

    Lisaius, M.C., Blake, A., Keshav, S., Atzberger, C.: Using barlow twins to create representations from cloud-corrupted remote sensing time series. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing17, 13162– 13168 (2024)

    2024

  22. [30]

    arXiv preprint arXiv:2510.26653 (2025) Warping EO for MIZ Labelling 17

    Martin, D., Gallego, J.: Towards reliable sea ice drift estimation in the arctic deep learning optical flow on radarsat-2. arXiv preprint arXiv:2510.26653 (2025) Warping EO for MIZ Labelling 17

    2025

  23. [31]

    The Cryosphere10(2), 913–925 (2016)

    Muckenhuber, S., Korosov, A.A., Sandven, S.: Open-source feature-tracking algo- rithm for sea ice drift retrieval from sentinel-1 sar imagery. The Cryosphere10(2), 913–925 (2016)

    2016

  24. [32]

    Remote Sensing16(20), 3838 (2024)

    Najem, S., Baghdadi, N., Bazzi, H., Zribi, M.: Incidence angle normalization of c-band radar backscattering coefficient over agricultural surfaces using dynamic cosine method. Remote Sensing16(20), 3838 (2024)

    2024

  25. [33]

    Communica- tions Earth & Environment6(1), 131 (2025)

    Raphael, M.N., Maierhofer, T.J., Fogt, R.L., Hobbs, W.R., Handcock, M.S.: A twenty-first century structural change in antarctica’s sea ice system. Communica- tions Earth & Environment6(1), 131 (2025)

    2025

  26. [34]

    Nature Geoscience14(11), 832–836 (2021)

    Riihelä, A., Bright, R.M., Anttila, K.: Recent strengthening of snow and ice albedo feedback driven by antarctic sea-ice loss. Nature Geoscience14(11), 832–836 (2021)

    2021

  27. [35]

    Remote Sensing of Environment305, 114073 (2024)

    Rogers, M.S., Fox, M., Fleming, A., van Zeeland, L., Wilkinson, J., Hosking, J.S.: Sea ice detection using concurrent multispectral and synthetic aperture radar im- agery. Remote Sensing of Environment305, 114073 (2024)

    2024

  28. [36]

    In: Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, Oc- tober 5-9, 2015, proceedings, part III 18

    Ronneberger, O., Fischer, P., Brox, T.: U-net: Convolutional networks for biomed- ical image segmentation. In: Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, Oc- tober 5-9, 2015, proceedings, part III 18....

    2015

  29. [37]

    The Cryosphere18(8), 3471–3494 (2024)

    Stokholm, A., Buus-Hinkler, J., Wulf, T., Korosov, A., Saldo, R., Pedersen, L.T., Arthurs, D., Dragan, I., Modica, I., Pedro, J., et al.: The autoice challenge. The Cryosphere18(8), 3471–3494 (2024)

    2024

  30. [38]

    EGUsphere2023, 1–25 (2023)

    Stokholm, A., Kucik, A., Longépé, N., Hvidegaard, S.M.: Ai4seaice: Task separa- tion and multistage inference cnns for automatic sea ice concentration charting. EGUsphere2023, 1–25 (2023)

    2023

  31. [39]

    IEEE Transactions on Geoscience and Remote Sensing (2025)

    Szwarcman, D., Roy, S., Fraccaro, P., Gíslason, O.E., Blumenstiel, B., Ghosal, R., De Oliveira, P.H., de Sousa Almeida, J.L., Sedona, R., Kang, Y., et al.: Prithvi- eo-2.0: A versatile multi-temporal foundation model for earth observation applica- tions. IEEE Transactions on G...

    2025

  32. [40]

    Geophysical Research Letters 52(2), e2024GL112000 (2025)

    Uusinoka, M., Haapala, J., Polojärvi, A.: Deep learning-based optical flow in fine- scale deformation mapping of sea ice dynamics. Geophysical Research Letters 52(2), e2024GL112000 (2025)

    2025

  33. [41]

    In: Indicators of Cli- mate Change, pp

    Vihma, T., Uotila, P.: Antarctic sea ice and climate change. In: Indicators of Cli- mate Change, pp. 81–102. Elsevier (2026)

    2026

  34. [42]

    Remote Sensing of Environment337, 115252 (2026)

    Wulf, T., Buus-Hinkler, J., Singha, S., Dasgupta, N., Athanasiadis, A., Kreiner, M.B.: A decade of sea ice concentration retrieved from sentinel-1. Remote Sensing of Environment337, 115252 (2026)

    2026

  35. [43]

    The Cryosphere18(11), 5277–5300 (2024)

    Wulf, T., Buus-Hinkler, J., Singha, S., Shi, H., Kreiner, M.B.: Pan-arctic sea ice concentration from sar and passive microwave. The Cryosphere18(11), 5277–5300 (2024)

    2024

  36. [44]

    arXiv preprint arXiv:2403.15356 (2024)

    Xiong, Z., Wang, Y., Zhang, F., Stewart, A.J., Hanna, J., Borth, D., Papoutsis, I., Saux, B.L., Camps-Valls, G., Zhu, X.X.: Neural plasticity-inspired multimodal foundation model for earth observation. arXiv preprint arXiv:2403.15356 (2024)

    2024

  37. [45]

    In: 2021 3rd International Conference on Advances in Computer Technology, Information Science and Communication (CTISC)

    Yu, T., Zhang, J., Zhou, J.: Conditional gan with effective attention for sar-to- optical image translation. In: 2021 3rd International Conference on Advances in Computer Technology, Information Science and Communication (CTISC). pp. 7–

    2021

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.