SatMamba: Development of Foundation Models for Remote Sensing Imagery Using State Space Models

Chuc Man Duc; Hiromichi Fukui

REVIEW 3 major objections 5 minor 1 cited by

SatMamba: Development of Foundation Models for Remote Sensing Imagery Using State Space Models

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

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

Pith's one-line read Mamba in a masked autoencoder matches ViT on remote sensing tasks

desk verdict SatMamba is a legitimate first step—MAE pretraining with a Mamba backbone for high-res RGB—but the linear-scaling selling point rests on resource curves, not on any long-sequence experiment. read the letter →

arxiv 2502.00435 v1 pith:J52WKW7W submitted 2025-02-01 cs.CV

Chuc Man Duc , Hiromichi Fukui This is my paper · ORCID

classification cs.CV

keywords maskedautoencoderstatespacemodelsMambaremotesensingsemanticsegmentationbuildingdamageassessmentself-supervisedpretraininglinearcomplexity

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 tries to show that a state-space-model backbone, Mamba, can replace the Vision Transformer inside a masked autoencoder for remote sensing pretraining, with no loss of downstream accuracy and with computational cost that grows linearly rather than quadratically with input length. The authors pretrain SatMamba on 363,572 RGB images from the fMoW dataset and fine-tune it on two image-to-image tasks: land-cover segmentation on OpenEarthMap and building damage assessment on xBD. On both tasks the best SatMamba variant edges out the ViT-based baselines and the strong CNN baselines. The efficiency argument is conditional: at the $224\times224$ RGB resolution used in all experiments SatMamba costs about twice the compute of the ViT baseline, and the linear-scaling benefit only appears as sequence length grows.

What carries the argument

The central object is the multi-way Mamba block: a layer containing several Mamba blocks, one per scanning direction, whose outputs are merged. Mamba is a selective state space model that maps a sequence $x(t)$ to $y(t)$ through a hidden state $h(t)$ with $h'(t)=Ah(t)+Bx(t)$ and $y(t)=Ch(t)+Dx(t)$, discretized with a timescale parameter; the paper uses Mamba 2. Four scanning directions let the model build a 2D-aware representation despite the 1D sequential scan. The encoder applies masking to the input after patchification and before flattening, keeps the natural patch order, and optionally adds fixed positional encodings. This design is what connects the masked autoencoder objective to a linear-complexity sequence model.

What would settle it

Fine-tune SatMamba-B and ViTMAE-B on inputs with genuinely long sequences, for example $512\times512$ crops, 10-band multispectral images, or six-frame temporal stacks, and compare wall-clock training time, peak memory, and task accuracy. If SatMamba consumes more compute and memory at those sizes while only matching accuracy, the paper's central efficiency-and-performance claim fails.

Watch

Extended reading notes

Core claim

SatMamba is a masked autoencoder whose encoder and decoder are built from multi-way Mamba blocks instead of transformer blocks. Masking is applied to the patch grid before flattening, the flattened sequence keeps its natural spatial order, and each layer scans the tokens in up to four directions (row-major forward and backward, column-major forward and backward) before merging. With these design choices, SatMamba-B without positional encodings reaches 66.46% mean IoU on OpenEarthMap and an 80.55% overall F1 score on xBD, the best results among ResNet50, EfficientNet-B7, ViTMAE-B, and ViTMAE-L; it also achieves lower pretraining reconstruction loss than ViTMAE-B. The paper claims this makes Mamba a viable backbone for remote sensing foundation models, with linear computational scaling that should benefit long sequences such as multispectral, multitemporal, and hyperspectral inputs, while acknowledging higher initial cost at small input sizes.

Load-bearing premise

The load-bearing premise is that Mamba's linear-in-sequence-length complexity will translate into a practical efficiency advantage on the long, multiband, or multitemporal satellite inputs where ViT becomes expensive; the paper's own experiments only exercise $224\times224$ RGB images, where SatMamba actually costs more.

Editorial extensions

If this is right

SatMamba can serve as a pretraining backbone for high-resolution RGB remote sensing, matching or exceeding ViT-based masked autoencoders on semantic segmentation and building damage assessment.
Because its cost scales linearly with input length, SatMamba should scale better than ViT on the long input sequences typical of multispectral, multitemporal, and hyperspectral data.
Removing fixed positional encodings improves SatMamba's fine-tuned accuracy, indicating that the sequential state itself carries spatial order information.
The architecture can be adapted to other image domains, including medium-resolution, multispectral, and natural images, since it relies only on standard patchification and masking.
Pretrained Mamba-based encoders can be dropped into a UNet-style fine-tuning pipeline for image-to-image tasks.

Reading between the lines

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

Beyond the paper's experiments, the linear-scaling advantage remains extrapolation: every downstream test runs at $224\times224$ RGB, so a fair comparison at $512\times512$ crops, 10-band multispectral images, or six-frame temporal stacks is needed to confirm that SatMamba's efficiency claim holds where ViT becomes expensive.
The result that positional encodings hurt SatMamba suggests that for state-space vision models the scan order may already provide enough positional information, a hypothesis worth testing on natural images and other SSM-based vision architectures.
Because SatMamba preserves natural patch order, it cannot exploit the random-shuffle ordering that MAE uses as a form of augmentation; a variant that shuffles masked tokens might either widen or close the gap with ViTMAE, and the paper does not test it.
If the linear scaling holds at long sequence lengths, SatMamba could process full multispectral cubes and temporal stacks directly, avoiding the spectral-group patching and pooling tricks currently used to keep ViT sequence lengths manageable.

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.

Desk Editor's Note

read the letter

Worth a look, but keep expectations calibrated. The genuinely new thing here is the first combination, in the cited literature, of masked autoencoding with a Mamba backbone for high-resolution RGB remote sensing, including a four-scan-direction design and a positional-encoding ablation. The paper ships code, the ablations are honest, and the authors openly state that SatMamba costs twice as much as ViTMAE-B at 224x224 and that multispectral/multitemporal evaluation is future work. That transparency is real credit.

What the paper actually shows: SatMamba-B w/o pos edges out ViTMAE-B on OpenEarthMap mIoU (66.46 vs 65.62) and on xBD overall F1 (80.55 vs 79.45). Those are modest gains, and EfficientNet-B7 is within a hair of both on both tasks, which matches the emerging picture that CNN baselines are not obsolete. The pretraining loss curves also look plausible, with SatMamba slightly lower than ViTMAE-B after 800 epochs.

Now the soft spots, in proportion. The central efficiency claim—linear scaling, advantageous for long sequences—is not demonstrated by any downstream experiment. Every test uses 224x224 RGB, where SatMamba is slower and heavier. The linear-scaling evidence is a resource curve extrapolated beyond tested sizes. That is a real gap, but the paper does not hide it; it is flagged in the text and in the conclusion. The parameter mismatch is also real: SatMamba-B has 229.86M params vs 111.66M for ViTMAE-B, so the performance comparison is capacity-confounded. Single-run results with no error bars make it hard to know if the 0.8–1.1 point gaps are meaningful. These are fixable with parameter-matched or compute-matched comparisons, long-sequence tests (even synthetic), and repeated runs.

One thing the stress-test note gets slightly wrong: the architecture does have linear complexity as a built-in property, so the theoretical claim is not fabricated. The issue is that the practical benefit for remote sensing is unverified. I would not call this a load-bearing flaw; I would call it an overreach in the abstract and conclusion relative to the evidence. The body is more careful.

Who is this for? Researchers working on Mamba-based vision backbones or geospatial foundation models. It is a useful data point, not a breakthrough. A serious referee should see it, because the architecture combination is plausible and the empirical setup is reproducible enough to improve.

Recommendation: accept for peer review, but with a clear request for parameter-matched comparisons, long-sequence evidence, and uncertainty quantification before publication.

Referee Report

3 major / 5 minor

Summary. SatMamba proposes a masked-autoencoder pretraining architecture for remote sensing imagery in which the ViT encoder/decoder blocks are replaced by multi-directional Mamba (state space) blocks. The method is pretrained on fMoW RGB images at 224x224 resolution and evaluated on two image-to-image downstream tasks: semantic segmentation on OpenEarthMap and building damage assessment on xBD. The paper reports that SatMamba-B without positional encodings achieves an mIoU of 66.46% on OpenEarthMap and an F1 score of 80.55% on xBD, slightly outperforming ViTMAE-B (65.62% mIoU, 79.45% F1) and EfficientNet-B7 (66.34% mIoU, 80.08% F1). The authors argue that SatMamba's linear computational scaling makes it particularly advantageous for long input sequences such as multispectral and multitemporal remote sensing data, while acknowledging that at the tested 224x224 resolution it consumes about twice the compute and more memory than ViTMAE-B.

Significance. If the central claims were fully supported, the paper would make a useful contribution: it would demonstrate that a state-space-model backbone can be integrated into an MAE pretraining framework for remote sensing, with downstream performance competitive with ViT-based MAEs and CNNs, plus a theoretical linear-scaling advantage for long sequences. The paper has several strengths: it provides source code, it honestly reports the parameter increase and higher initial cost of SatMamba, it includes CNN baselines that are often missing in remote sensing foundation model papers, it performs ablations over scanning directions and positional encodings, and it evaluates on two distinct downstream tasks. The main weakness is that the experiments do not actually exercise the long-sequence regime that motivates the architecture, and the performance comparison is confounded by a 2x parameter/compute gap. The work is therefore incremental but potentially valuable if the claims are either strengthened with additional experiments or appropriately narrowed.

major comments (3)

[Section IV-A, Table I, Figure 3] The comparison is not parameter-matched. SatMamba-B has 229.86M parameters versus 111.66M for ViTMAE-B, and at the 224x224 resolution used in all experiments it consumes roughly twice the compute and more GPU memory than ViTMAE-B, as the paper itself states. The paper also notes that the increased parameter count was necessary to achieve pretraining results comparable to ViTMAE-B. Consequently, the reported downstream gains (Tables II and III) may reflect the additional capacity or compute rather than the Mamba architecture itself. A parameter-matched or compute-matched comparison is needed to support the claim that SatMamba is a viable alternative to ViT-based pretraining.
[Section III-A.1 and Section IV-A] The central motivating advantage, linear scaling for long remote-sensing sequences, is not tested. All pretraining and downstream experiments use 224x224 RGB images; the paper explicitly defers multispectral and multitemporal evaluation to future work. Figure 3 is a resource curve extrapolated to larger input sizes, not a measurement of downstream accuracy or speed at long sequence lengths. The conclusion that SatMamba 'enables efficient self-supervised learning' for such data is therefore unsupported. Either add a long-sequence or multiband experiment, or revise the conclusion to state that linear scaling is a theoretical property that remains to be demonstrated.
[Section IV-B, Tables II and III] Results are reported from single runs without variance estimates. The gaps between SatMamba-B w/o pos and ViTMAE-B are 0.84 mIoU points (66.46 vs 65.62) and 1.10 F1 points (80.55 vs 79.45). In the absence of multiple seeds or statistical testing, these small margins could be within run-to-run variability, which weakens the 'competitive' claim. Please report mean and standard deviation over at least three seeds, or provide a statistical significance test.

minor comments (5)

[Section III-A.2] The text says the tensor has shape R(H/P)x(H/P)xD, but it should presumably be R(H/P)x(W/P)xD; the width dimension is missing.
[Figure 1] Figure 1 is never explicitly referenced in the method text; please add a reference when the encoder and decoder are first described.
[Author block] The author affiliation sentence reads 'Hiromichi Fukui are with'; it should be 'Hiromichi Fukui is with'.
[Section IV-B.2] The evaluation protocol is attributed to reference [11], a change detection paper; please verify that this is the correct source for the xBD evaluation metrics.
[Section IV-B.1] The discussion of ViTMAE-B outperforming ViTMAE-L mentions pretraining resolution and model size as possible causes, but this confound is not controlled; it would be helpful to acknowledge this limitation in the discussion.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found; the central claims rest on external benchmarks and architectural properties rather than on fitted inputs or self-citation.

full rationale

We reviewed the derivation chain and found no circular step that reduces a claimed result to its own input. The SatMamba design choices (state dimension 64, head dimension 96, four scanning directions, roughly twice the parameters of ViTMAE-B) are presented as ablations and engineering adjustments to match pretraining loss, not as fitted constants that are later relabeled as predictions. The downstream claims are evaluated on external benchmarks (OpenEarthMap and xBD) with IoU and F1 metrics, which are not derivable from the pretraining loss or from the model configuration itself; in fact, the paper explicitly notes that pretraining loss need not transfer linearly to fine-tuning loss. The linear-complexity claim is a stated property of the Mamba recurrence (Eqs. 1-2) supported by resource curves in Figure 3, not a renamed empirical pattern or a parameter fitted to the target result. Citations to SatMAE, MAE, Mamba, Vision Mamba, and VMamba are external prior works and are not used as load-bearing justification for SatMamba's own experimental outcomes. The identified weaknesses, such as parameter-count mismatch between SatMamba-B (229.86M) and ViTMAE-B (111.66M) and the absence of long-sequence or multispectral experiments, are correctness and generalization concerns, not circularity.

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

The central claim rests on Mamba architecture properties and hyperparameters chosen by hand; no new physical entities are introduced.

free parameters (4)

Scanning directions = 4 (row/col, fwd/bwd)
Number of Mamba blocks per multi-way layer; ablation in Section IV-A shows full scanning gives the lowest pretraining loss.
Inner state dimension = 64
Mamba-specific parameter set to 64 to align with ViTMAE-B compute budget (Section IV-A).
Head dimension = 96
Set to 96 along with 4-way blocks to reach about 2x ViTMAE-B parameter count (Section IV-A).
Masking ratio = 0.75
Follows standard MAE 75% masking (Section IV-A).

assumptions (4)

standard math Mamba's selective state space model and zero-order hold discretization correctly model image sequences.
Equations (1)-(2) are taken from [7] as given; no verification is provided in this paper.
domain assumption Linear computational scaling holds for the multi-way Mamba implementation with four scanning directions.
Figure 3 plots compute and memory vs input size, but no downstream experiments at large sequence lengths validate the efficiency claim.
domain assumption Pretrained representations transfer from fMoW RGB 224x224 images to OpenEarthMap and xBD with a UNet decoder.
Fine-tuning protocol in Section III-B assumes transfer across datasets and resolutions without explicit validation.
ad hoc to paper Four scanning directions plus normalized-pixel reconstruction provide a positional-encoding-free inductive bias sufficient for spatial understanding.
Design choice in Section III-A-2 made because Mamba is sequence-based; the w/o pos variant wins empirically, but the mechanism is not analyzed.

how reviews work

0 comments

Cite this review

Pith. "Pith review of SatMamba: Development of Foundation Models for Remote Sensing Imagery Using State Space Models." pith.science (2026). https://pith.science/paper/J52WKW7W

@misc{pith2026250200435,
  author       = {Pith},
  title        = {Pith review of: SatMamba: Development of Foundation Models for Remote Sensing Imagery Using State Space Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/J52WKW7W}},
  note         = {Machine review of arXiv:2502.00435}
}

read the original abstract

Foundation models refer to deep learning models pretrained on large unlabeled datasets through self-supervised algorithms. In the Earth science and remote sensing communities, there is growing interest in transforming the use of Earth observation data, including satellite and aerial imagery, through foundation models. Various foundation models have been developed for remote sensing, such as those for multispectral, high-resolution, and hyperspectral images, and have demonstrated superior performance on various downstream tasks compared to traditional supervised models. These models are evolving rapidly, with capabilities to handle multispectral, multitemporal, and multisensor data. Most studies use masked autoencoders in combination with Vision Transformers (ViTs) as the backbone for pretraining. While the models showed promising performance, ViTs face challenges, such as quadratic computational scaling with input length, which may limit performance on multiband and multitemporal data with long sequences. This research aims to address these challenges by proposing SatMamba, a new pretraining framework that combines masked autoencoders with State Space Model, offering linear computational scaling. Experiments on high-resolution imagery across various downstream tasks show promising results, paving the way for more efficient foundation models and unlocking the full potential of Earth observation data. The source code is available in https://github.com/mdchuc/HRSFM.

Figures

Figures reproduced from arXiv: 2502.00435 by the authors.

**Figure 3.** Resource requirements of different models across varying input sizes: (a) Computational requirements; (b) Memory requirements. important to note that these results are meant for baseline comparison, highlighting that the two models exhibit comparable pretraining performance. However, it is not necessary that the pretraining loss transfers linearly to the fine-tuning loss. The computational cost of SatMamba increas… view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Vision Mamba in Remote Sensing: A Comprehensive Survey of Techniques, Applications and Outlook
cs.CV 2025-05 conditional novelty 5.0 of 10

A structured review of roughly 120 Mamba-based remote sensing papers that proposes taxonomies for scan strategies and architectural integrations, and claims Mamba-based models often outperform CNN and Transformer base...

Reference graph

Works this paper leans on

23 extracted references · 17 canonical work pages · cited by 1 Pith paper

[1]

SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery,

Y . Cong, S. Khanna, C. Meng, P. Liu, E. Rozi, Y . He, M. Burke, D. B. Lobell, and S. Ermon, “SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery,” Advances in Neural Information Processing Systems , vol. 35, no. NeurIPS, pp. 1–24, 2022

work page 2022
[2]

Foundation Models for Generalist Geospatial Artificial Intelligence,

J. Jakubik, S. Roy, C. E. Phillips, P. Fraccaro, D. Godwin, B. Zadrozny, D. Szwarcman, C. Gomes, G. Nyirjesy, B. Edwards, D. Kimura, N. Si- mumba, L. Chu, S. K. Mukkavilli, D. Lambhate, K. Das, R. Bangalore, D. Oliveira, M. Muszynski, K. Ankur, M. Ramasubramanian, I. Gurung, S. Khallaghi, H. S. Li, M. Cecil, M. Ahmadi, F. Kordi, H. Alemo- hammad, M. Maske...

work page 2023
[3]

SpectralGPT: Spectral Remote Sensing Foundation Model,

D. Hong, B. Zhang, X. Li, Y . Li, C. Li, J. Yao, N. Yokoya, H. Li, X. Jia, A. Plaza, G. Paolo, J. A. Benediktsson, and J. Chanussot, “SpectralGPT: Spectral Remote Sensing Foundation Model,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 46, no. 8, pp. 5227–5244, 2024. [Online]. Available: 10.1109/TPAMI.2024.3362475

arXiv 2024
[4]

Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observation,

Z. Xiong, Y . Wang, F. Zhang, A. J. Stewart, D. Borth, I. Papoutsis, B. L. Saux, G. Camps-valls, and X. X. Zhu, “Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observation,” arXiv, pp. 1–36, 2024

work page 2024
[5]

Masked Autoencoders Are Scalable Vision Learners,

K. He, X. Chen, S. Xie, Y . Li, P. Dollar, and R. Girshick, “Masked Autoencoders Are Scalable Vision Learners,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recog- nition, vol. 2022-June, pp. 15 979–15 988, 2022

work page 2022
[6]

An Image Is Worth 16X16 Words: Transformers for Image Recognition At Scale,

A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, and N. Houlsby, “An Image Is Worth 16X16 Words: Transformers for Image Recognition At Scale,” ICLR 2021 - 9th International Conference on Learning Representations , 2021

work page 2021
[7]

Mamba: Linear-Time Sequence Modeling with Selective State Spaces,

A. Gu and T. Dao, “Mamba: Linear-Time Sequence Modeling with Selective State Spaces,” arXiv, pp. 1–36, 2023. [Online]. Available: http://arxiv.org/abs/2312.00752

arXiv 2023
[8]

VM-UNet: Vision Mamba UNet for Medical Image Segmentation

J. Ruan and S. Xiang, “VM-UNet: Vision Mamba UNet for Medical Image Segmentation.”

work page

Show all 23 references

[9]

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

L. Zhu, B. Liao, Q. Zhang, X. Wang, W. Liu, and X. Wang, “Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model,” Proceedings of the 41 st International Conference on Machine Learning , vol. 235, pp. 62 429–62 442, 2024

2024
[10]

VMamba: Visual State Space Model,

Y . Liu, Y . Tian, Y . Zhao, H. Yu, L. Xie, Y . Wang, Q. Ye, and Y . Liu, “VMamba: Visual State Space Model,” arXiv, 2024. [Online]. Available: http://arxiv.org/abs/2401.10166

2024 arXiv
[11]

ChangeMamba: Remote Sensing Change Detection with Spatiotemporal State Space Model,

H. Chen, J. Song, C. Han, J. Xia, and N. Yokoya, “ChangeMamba: Remote Sensing Change Detection with Spatiotemporal State Space Model,” IEEE Transactions on Geoscience and Remote Sensing , vol. 62, pp. 1–19, 2024

2024
[12]

Revisiting pre-trained remote sensing model benchmarks: resizing and normalization matters,

I. Corley, C. Robinson, R. Dodhia, J. M. L. Ferres, and P. Najafirad, “Revisiting pre-trained remote sensing model benchmarks: resizing and normalization matters,” 2023. [Online]. Available: http://arxiv.org/abs/ 2305.13456

2023 arXiv
[13]

PhilEO Bench: Evaluating Geo-Spatial Foundation Models,

C. Fibaek, L. Camilleri, A. Luyts, N. Dionelis, and B. L. Saux, “PhilEO Bench: Evaluating Geo-Spatial Foundation Models,” pp. 3–8,
[14]

Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality,

T. Dao and A. Gu, “Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality,” pp. 1–52, 2024. [Online]. Available: http://arxiv.org/abs/2405.21060

2024 arXiv
[15]

U-net: Convolutional networks for biomedical image segmentation,

O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” in MICCAI, 2015

2015
[16]

Segmentation Models Pytorch,

P. Iakubovskii, “Segmentation Models Pytorch,” GitHub repository ,
[17]

Prithvi Pytorch,

I. Corley, “Prithvi Pytorch,” GitHub repository , 2024. [Online]. Available: https://github.com/isaaccorley/prithvi-pytorch

2024
[18]

Creating xbd: A dataset for assessing building damage from satellite imagery,

R. Gupta, B. Goodman, N. Patel, R. Hosfelt, S. Sajeev, E. Heim, J. Doshi, K. Lucas, H. Choset, and M. Gaston, “Creating xbd: A dataset for assessing building damage from satellite imagery,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (C...

2019
[19]

Functional Map of the World,

G. Christie, N. Fendley, J. Wilson, and R. Mukherjee, “Functional Map of the World,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition , vol. 2, pp. 6172–6180, 2018

2018
[20]

USat: A Unified Self-Supervised Encoder for Multi-Sensor Satellite Imagery,

J. Irvin, L. Tao, J. Zhou, Y . Ma, L. Nashold, B. Liu, and A. Y . Ng, “USat: A Unified Self-Supervised Encoder for Multi-Sensor Satellite Imagery,” arXiv, 2023. [Online]. Available: http://arxiv.org/abs/2312.02199

2023 arXiv
[21]

OpenEarthMap: A Benchmark Dataset for Global High-Resolution Land Cover Map- ping,

J. Xia, N. Yokoya, B. Adriano, and C. Broni-Bediako, “OpenEarthMap: A Benchmark Dataset for Global High-Resolution Land Cover Map- ping,” Proceedings - 2023 IEEE Winter Conference on Applications of Computer Vision, WACV 2023 , pp. 6243–6253, 2023

2023
[2019]

Available: https://github.com/qubvel/segmentation models.pytorch

[Online]. Available: https://github.com/qubvel/segmentation models.pytorch
[2024]

Available: http://arxiv.org/abs/2401.04464

[Online]. Available: http://arxiv.org/abs/2401.04464

arXiv

Pith tools

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

[1] [1]

SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery,

Y . Cong, S. Khanna, C. Meng, P. Liu, E. Rozi, Y . He, M. Burke, D. B. Lobell, and S. Ermon, “SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery,” Advances in Neural Information Processing Systems , vol. 35, no. NeurIPS, pp. 1–24, 2022

work page 2022

[2] [2]

Foundation Models for Generalist Geospatial Artificial Intelligence,

J. Jakubik, S. Roy, C. E. Phillips, P. Fraccaro, D. Godwin, B. Zadrozny, D. Szwarcman, C. Gomes, G. Nyirjesy, B. Edwards, D. Kimura, N. Si- mumba, L. Chu, S. K. Mukkavilli, D. Lambhate, K. Das, R. Bangalore, D. Oliveira, M. Muszynski, K. Ankur, M. Ramasubramanian, I. Gurung, S. Khallaghi, H. S. Li, M. Cecil, M. Ahmadi, F. Kordi, H. Alemo- hammad, M. Maske...

work page 2023

[3] [3]

SpectralGPT: Spectral Remote Sensing Foundation Model,

D. Hong, B. Zhang, X. Li, Y . Li, C. Li, J. Yao, N. Yokoya, H. Li, X. Jia, A. Plaza, G. Paolo, J. A. Benediktsson, and J. Chanussot, “SpectralGPT: Spectral Remote Sensing Foundation Model,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 46, no. 8, pp. 5227–5244, 2024. [Online]. Available: 10.1109/TPAMI.2024.3362475

arXiv 2024

[4] [4]

Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observation,

Z. Xiong, Y . Wang, F. Zhang, A. J. Stewart, D. Borth, I. Papoutsis, B. L. Saux, G. Camps-valls, and X. X. Zhu, “Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observation,” arXiv, pp. 1–36, 2024

work page 2024

[5] [5]

Masked Autoencoders Are Scalable Vision Learners,

K. He, X. Chen, S. Xie, Y . Li, P. Dollar, and R. Girshick, “Masked Autoencoders Are Scalable Vision Learners,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recog- nition, vol. 2022-June, pp. 15 979–15 988, 2022

work page 2022

[6] [6]

An Image Is Worth 16X16 Words: Transformers for Image Recognition At Scale,

A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, and N. Houlsby, “An Image Is Worth 16X16 Words: Transformers for Image Recognition At Scale,” ICLR 2021 - 9th International Conference on Learning Representations , 2021

work page 2021

[7] [7]

Mamba: Linear-Time Sequence Modeling with Selective State Spaces,

A. Gu and T. Dao, “Mamba: Linear-Time Sequence Modeling with Selective State Spaces,” arXiv, pp. 1–36, 2023. [Online]. Available: http://arxiv.org/abs/2312.00752

arXiv 2023

[8] [8]

VM-UNet: Vision Mamba UNet for Medical Image Segmentation

J. Ruan and S. Xiang, “VM-UNet: Vision Mamba UNet for Medical Image Segmentation.”

work page