Pith. sign in

REVIEW 3 major objections 5 minor 37 references

Few-Shot Adaptation Benchmark for Remote Sensing Vision-Language Models

T0 review · 3 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

Pith's one-line read This paper shows that zero-shot accuracy does not predict few-shot adaptation for remote sensing vision-language models, and establishes the first structured benchmark that exposes the gap.

desk verdict Useful first few-shot adaptation benchmark for RSVLMs, but the unverified pretraining-disjointness premise could flip the main ranking. read the letter →

arxiv 2510.07135 v1 pith:VN3RU2JR submitted 2025-10-08 cs.CV

classification cs.CV
keywords few-shotlearningvision-languagemodelsremotesensingsceneclassificationbenchmarkCLIPprompttuninglow-rankadaptation
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 the remote sensing community cannot infer few-shot adaptation ability from zero-shot accuracy, and backs that claim with the first structured benchmark dedicated to low-data adaptation of remote sensing vision-language models. Across ten scene-classification datasets, three RS-specialized models, and five adaptation methods, it finds that GeoRSCLIP consistently adapts best, while the best adaptation method shifts with shot count: CLIP-LoRA leads at two to four shots, Tip-Adapter at sixteen. The paper also shows that low-rank finetuning scales better to large backbones than prompt-tuning or text-residual methods. A reproducible, extensible codebase accompanies the benchmark, so future models and methods can be compared under identical conditions.

What carries the argument

The central object is the benchmark suite itself: ten remote sensing scene-classification datasets, three RS-specialized vision-language models (RemoteCLIP, GeoRSCLIP, SkyCLIP) plus the original CLIP, and five few-shot adaptation methods (CoOp, MaPLe, TaskRes, Tip-Adapter, CLIP-LoRA). The comparisons are carried by a controlled protocol: fixed random train/validation/test splits, three seeds, a shared ViT-B/32 backbone for the main model comparison, and the original hyperparameters for each adaptation method. This setup isolates the influence of pretraining data and adaptation strategy from architectural differences, making the observed ranking and shot-dependent trends interpretable.

What would settle it

A concrete test: audit the public pretraining corpora (RS5M, RET-3, SkyScript) for image or class overlap with the ten benchmark datasets, or re-run the benchmark on held-out datasets collected after the models' release; if GeoRSCLIP's margin shrinks or vanishes on uncontaminated data, the claim that it is inherently more amenable to few-shot adaptation would be weakened.

Watch

Extended reading notes

Core claim

The central finding is that zero-shot performance is not a reliable proxy for few-shot adaptation in remote sensing scene classification. Models with nearly equal zero-shot accuracy, such as GeoRSCLIP and SkyCLIP, diverge once trained on a handful of labeled examples, with GeoRSCLIP ahead across all five adaptation methods on a matched backbone. At the method level, no single strategy dominates across datasets and shot counts: CLIP-LoRA is strongest on average at 2-4 shots, Tip-Adapter becomes best at 16 shots, TaskRes excels on high-class-count datasets, and MaPLe lags. Scaling experiments reinforce the low-rank result: CLIP-LoRA remains accurate on ViT-L/14 and ViT-H/14 backbones while CoO

Load-bearing premise

The load-bearing premise is that none of the ten evaluation datasets were used in pretraining the implemented RSVLMs; the paper asserts this in Section II-A but does not audit the pretraining corpora, and if any benchmark dataset leaked into training the zero-shot and few-shot numbers would be inflated and the cross-model comparison contaminated.

Editorial extensions

If this is right

  • GeoRSCLIP should be the default RSVLM for few-shot scene classification when labels are scarce, since it led across all five adaptation methods on the shared backbone.
  • Ranking adaptation methods by a single shot count is unreliable; any practical recommendation should cite the supervision level, because CLIP-LoRA leads at low shots while Tip-Adapter wins at 16.
  • Low-rank finetuning is the safest choice when scaling to large backbones: CLIP-LoRA kept improving with backbone size, while CoOp, MaPLe, and TaskRes lost accuracy on ViT-H/14.
  • Even one labeled example per class produces a significant improvement over zero-shot evaluation across all models and methods, so minimal supervision is worth exploiting.
  • The benchmark's open codebase allows future RSVLMs and adaptation methods to be inserted and compared under the same protocol, turning few-shot performance into a checkable quantity rather than an assumption.

Reading between the lines

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

  • If zero-shot and few-shot rankings diverge as broadly as this benchmark suggests, zero-shot leaderboards for RSVLMs should be accompanied by a few-shot evaluation before deployment decisions are made; the paper demonstrates the gap but stops short of prescribing a standard few-shot reporting protocol.
  • The dataset-dependent method rankings hint that the optimal tuning location—text embeddings, prompts, intermediate weights, or cached features—depends on class granularity and class count; TaskRes's strength on MLRSNet and RESISC45 suggests a testable hypothesis that text-side residual tuning helps when many fine-grained classes share visual structure.
  • A natural extension is to measure whether the GeoRSCLIP advantage persists under distribution shift or on other RS tasks such as segmentation and object detection, not just scene classification; the paper's fixed-seed, three-run protocol makes such extensions directly comparable.
  • The lack of an overlap audit between the RSVLMs' pretraining corpora and the benchmark datasets means the reported margins should be re-examined if any overlap is found; the paper's own numbers would remain internally comparable but their absolute size could be inflated.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper introduces a benchmark for few-shot adaptation of remote sensing vision-language models (RSVLMs). It evaluates four models (GeoRSCLIP, RemoteCLIP, SkyCLIP, and CLIP) on ten RS scene-classification datasets, using five adaptation methods (CoOp, MaPLe, TaskRes, Tip-Adapter, CLIP-LoRA). Results are reported for 0, 1, 2, 4, 8, and 16 shots, mostly as averages over three seeds, with detailed tables for GeoRSCLIP and additional scaling experiments on larger backbones. The central claims are that GeoRSCLIP consistently outperforms the other models under all adaptation methods; that zero-shot performance is not a reliable predictor of few-shot adaptation performance; and that no single adaptation method dominates, with CLIP-LoRA strongest at low shots and Tip-Adapter strongest at 16 shots. The paper also promises open-source code.

Significance. If the empirical findings hold, this is a useful first structured benchmark for few-shot adaptation of RSVLMs. The multi-dataset scope, the inclusion of several adaptation families, and the promise of reproducible code are strengths. The finding that method rankings are shot- and dataset-dependent would be practically valuable. However, the main conclusions currently rest on an unverified pretraining-disjointness claim and on three-seed averages without any noise characterization, so the significance of the specific rankings is not yet established.

major comments (3)
  1. [II-A] The load-bearing premise that 'none of the selected evaluation datasets were used during the pretraining of the implemented RSVLMs' is asserted but never audited. GeoRSCLIP is fine-tuned on RS5M, RemoteCLIP on RET-3/SEG-4/DET-10, and SkyCLIP on SkyScript; these public corpora are often assembled from existing RS scene datasets, so overlap with EuroSAT, RESISC45, AID, PatternNet, MLRSNet, and similar is plausible. If any benchmark test images or near-duplicates appeared during pretraining, the zero-shot and few-shot accuracies in Figure 1 and Tables III–V are inflated for the affected models, directly contaminating the paper's central cross-model ranking. Please provide an overlap audit (e.g., URL/ID filtering, duplicate/near-duplicate detection, or dataset-construction documentation) and re-run or qualify the results accordingly.
  2. [III (Fig. 1, Tables III–V)] All quantitative claims rest on averages over three random seeds, yet no standard deviations, confidence intervals, or significance tests are reported anywhere. For example, Table III shows differences of 1–3 points between methods (e.g., 4-shot TaskRes average 85.4 vs. CLIP-LoRA 87.8; 1-shot TaskRes 79.7 vs. CLIP-LoRA 79.4), and Figure 1 has no error bars. Without per-seed variability, the assertions that 'GeoRSCLIP consistently outperforms' and that 'zero-shot performance is not always a reliable indicator' are not statistically supported. Please report per-seed results or standard deviations, and state the exact support-set generation procedure: the seed values, whether support sets are class-balanced, and whether the same support sets are used across methods and models.
  3. [II-C / IV] The paper's central methodological claim is that it provides a reproducible benchmark, but the description of the few-shot protocol is incomplete. Section II-C defines the support set formally but does not specify how the C×K examples are sampled per seed, what the query set is, or whether validation splits are used for early stopping or hyperparameter selection. The split '50/25/25 with a fixed random seed' is mentioned in II-A, but the relationship between that seed and the 'three random seeds' used in the experiments is unclear. Without these details, the benchmark is not fully reproducible and the fairness of the method comparison is difficult to assess. Please specify the protocol precisely and, ideally, release the exact splits and support-set indices.
minor comments (5)
  1. [Abstract/Sec. IV] The GitHub URL in the full text and code-accessibility section contains a space ('fewshot RSVLMs') and should be a proper hyperlink with the underscore; the abstract version is correct.
  2. [Fig. 1] The axis labels contain repeated placeholder text '(averaged out on 10 datasets)' and lack error bars; this makes the figure hard to parse and overstates certainty.
  3. [Table II] The 'Vision Encoder Backbones' row is difficult to read because column boundaries are not clear. Please format the backbone lists explicitly for each model.
  4. [Table I] For unbalanced datasets (AID, MLRSNet, RSICB128, RSICB256) the 'Available Training Samples per Class (avg.)' is an average over classes and could be misleading; consider giving the per-class range or median.
  5. [Throughout] The name 'Tip-Adapter' is inconsistently capitalized as 'TIP-Adapter' in several places (e.g., Table IV, Table V). Please standardize.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper is a direct empirical benchmark using externally defined models and adaptation methods, with no fitted quantity renamed as a prediction.

full rationale

The paper's claims are empirical measurements, not derivations. The benchmark compares three published RSVLMs and five published few-shot adaptation methods under fixed splits and the original hyperparameters from their respective publications (Section II-C: 'For each method, we adopt the original hyperparameter settings specified in their respective publication'). No parameter is fitted to the benchmark and then reported as a prediction; the central observations — GeoRSCLIP's consistent advantage and the dissociation between zero-shot and few-shot rankings — are read directly from Tables III–V and Figure 1. The inclusion of CLIP-LoRA, whose paper shares a co-author, is not used as load-bearing evidence; it is one of five externally defined methods evaluated against the others, and the paper also reports its higher computational cost. Self-citations in the references are contextual and do not supply the conclusions. The one load-bearing factual premise, in Section II-A, is that 'none of the selected evaluation datasets were used during the pretraining of the implemented RSVLMs'; this is asserted without an overlap audit and could affect validity if false, but it is an external-contamination risk, not a definitional or fitted-input circularity. It does not reduce any reported result to an input by construction. Therefore the appropriate finding is no significant circularity.

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

No free parameters are fit: the paper evaluates published methods with their original hyperparameters. Its conclusions depend on dataset/model/method selection and on the unverified pretraining-disjointness assumption, not on equations, so the ledger is dominated by domain assumptions rather than numeric inputs.

assumptions (4)
  • domain assumption None of the ten benchmark datasets were used during pretraining of RemoteCLIP, GeoRSCLIP, or SkyCLIP.
    Stated without audit in §II-A; public descriptions of RS5M and RET-3 suggest they aggregate existing RS datasets.
  • domain assumption Default hyperparameters from original publications of CoOp, MaPLe, TaskRes, Tip-Adapter, CLIP-LoRA transfer appropriately to RS.
    Adopted in §II-C; different backbone sizes and RS domain may require tuning, which is not explored.
  • domain assumption Three random seeds are sufficient to characterize few-shot performance.
    Section III uses three seeds but reports no variance; few-shot support sets can be high-variance.
  • domain assumption Comparing all models with ViT-B/32 isolates pretraining differences.
    Section III-A; assumes training dynamics and implementation are equalized across models.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Few-Shot Adaptation Benchmark for Remote Sensing Vision-Language Models." pith.science (2026). https://pith.science/paper/VN3RU2JR

@misc{pith2026251007135,
  author       = {Pith},
  title        = {Pith review of: Few-Shot Adaptation Benchmark for Remote Sensing Vision-Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VN3RU2JR}},
  note         = {Machine review of arXiv:2510.07135}
}
read the original abstract

Remote Sensing Vision-Language Models (RSVLMs) have shown remarkable potential thanks to large-scale pretraining, achieving strong zero-shot performance on various tasks. However, their ability to generalize in low-data regimes, such as few-shot learning, remains insufficiently explored. In this work, we present the first structured benchmark for evaluating few-shot adaptation methods on RSVLMs. We conduct comprehensive experiments across ten remote sensing scene classification datasets, applying five widely used few-shot adaptation strategies to three state-of-the-art RSVLMs with varying backbones. Our findings reveal that models with similar zero-shot performance can exhibit markedly different behavior under few-shot adaptation, with some RSVLMs being inherently more amenable to such adaptation than others. The variability of performance and the absence of a clear winner among existing methods highlight the need for the development of more robust methods for few-shot adaptation tailored to RS. To facilitate future research, we provide a reproducible benchmarking framework and open-source code to systematically evaluate RSVLMs under few-shot conditions. The source code is publicly available on Github: https://github.com/elkhouryk/fewshot_RSVLMs

Figures

Figures reproduced from arXiv: 2510.07135 by the authors.

Figure 1
Figure 1. Performance evaluation of four vision-language models ( [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Comparison of training time for five few-shot adaptation methods [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 1 linked inside Pith

  1. [1]

    Deep learning in environmental remote sensing: Achievements and challenges,

    Q. Yuan, H. Shen, T. Li, Z. Li, S. Li, Y . Jiang, H. Xu, W. Tan, Q. Yang, J. Wanget al., “Deep learning in environmental remote sensing: Achievements and challenges,”Remote sensing of Environment, vol. 241, p. 111716, 2020

  2. [2]

    From satellite to uav-based remote sensing: A review on precision agriculture,

    S. K. Phang, T. H. A. Chiang, A. Happonen, and M. M. L. Chang, “From satellite to uav-based remote sensing: A review on precision agriculture,” Ieee Access, vol. 11, pp. 127 057–127 076, 2023

  3. [3]

    Streamlined hybrid annotation framework using scalable codestream for bandwidth-restricted uav object detection,

    K. El Khoury, T. Godelaine, S. Delvaux, S. Lugan, and B. Macq, “Streamlined hybrid annotation framework using scalable codestream for bandwidth-restricted uav object detection,” in2024 IEEE International Conference on Image Processing (ICIP), 2024, pp. 1581–1587

  4. [4]

    Remote sensing image scene classification meets deep learning: Challenges, methods, benchmarks, and opportunities,

    G. Cheng, X. Xie, J. Han, L. Guo, and G.-S. Xia, “Remote sensing image scene classification meets deep learning: Challenges, methods, benchmarks, and opportunities,”IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 13, pp. 3735– 3756, 2020

  5. [5]

    Semi-supervised remote-sensing image scene classification using representation consistency siamese network,

    W. Miao, J. Geng, and W. Jiang, “Semi-supervised remote-sensing image scene classification using representation consistency siamese network,” IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1– 14, 2022

  6. [6]

    Stacked autoencoders driven by semi-supervised learning for building extraction from near infrared remote sensing imagery,

    E. Protopapadakis, A. Doulamis, N. Doulamis, and E. Maltezos, “Stacked autoencoders driven by semi-supervised learning for building extraction from near infrared remote sensing imagery,”Remote Sensing, vol. 13, no. 3, p. 371, 2021

  7. [7]

    Semiroadexnet: A semi- supervised network for road extraction from remote sensing imagery via adversarial learning,

    H. Chen, Z. Li, J. Wu, W. Xiong, and C. Du, “Semiroadexnet: A semi- supervised network for road extraction from remote sensing imagery via adversarial learning,”ISPRS Journal of Photogrammetry and Remote Sensing, vol. 198, pp. 169–183, 2023

  8. [8]

    Few-shot remote sensing image scene classification: Recent advances, new baselines, and future trends,

    C. Qiu, X. Zhang, X. Tong, N. Guan, X. Yi, K. Yang, J. Zhu, and A. Yu, “Few-shot remote sensing image scene classification: Recent advances, new baselines, and future trends,”ISPRS Journal of Photogrammetry and Remote Sensing, vol. 209, pp. 368–382, 2024

Show all 37 references
  1. [9]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clarket al., “Learning transferable visual models from natural language supervision,” inInternational conference on machine learning. PmLR, 2021, pp. 8748–8763

  2. [10]

    Rs5m and georsclip: A large scale vision-language dataset and a large vision-language model for remote sensing,

    Z. Zhang, T. Zhao, Y . Guo, and J. Yin, “Rs5m and georsclip: A large scale vision-language dataset and a large vision-language model for remote sensing,”IEEE Transactions on Geoscience and Remote Sensing, 2024

  3. [11]

    Remoteclip: A vision language foundation model for remote sensing,

    F. Liu, D. Chen, Z. Guan, X. Zhou, J. Zhu, Q. Ye, L. Fu, and J. Zhou, “Remoteclip: A vision language foundation model for remote sensing,” IEEE Transactions on Geoscience and Remote Sensing, 2024

  4. [12]

    Skyscript: A large and semantically diverse vision-language dataset for remote sens- ing,

    Z. Wang, R. Prabha, T. Huang, J. Wu, and R. Rajagopal, “Skyscript: A large and semantically diverse vision-language dataset for remote sens- ing,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 6, 2024, pp. 5805–5813

  5. [13]

    H2rsvlm: Towards helpful and honest remote sensing large vision language model,

    C. Pang, J. Wu, J. Li, Y . Liu, J. Sun, W. Li, X. Weng, S. Wang, L. Feng, G.-S. Xiaet al., “H2rsvlm: Towards helpful and honest remote sensing large vision language model,”arXiv e-prints, pp. arXiv–2403, 2024

  6. [14]

    Lhrs-bot: Empowering remote sensing with vgi-enhanced large multimodal language model,

    D. Muhtar, Z. Li, F. Gu, X. Zhang, and P. Xiao, “Lhrs-bot: Empowering remote sensing with vgi-enhanced large multimodal language model,” inEuropean Conference on Computer Vision. Springer, 2024, pp. 440– 457

  7. [15]

    Enhancing remote sensing vision-language models for zero-shot scene classification,

    K. El Khoury, M. Zanella, B. G ´erin, T. Godelaine, B. Macq, S. Mah- moudi, C. De Vleeschouwer, and I. Ben Ayed, “Enhancing remote sensing vision-language models for zero-shot scene classification,” in ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Sig...

  8. [16]

    Learning to prompt for vision- language models,

    K. Zhou, J. Yang, C. C. Loy, and Z. Liu, “Learning to prompt for vision- language models,”International Journal of Computer Vision, vol. 130, no. 9, pp. 2337–2348, 2022

  9. [17]

    Maple: Multi-modal prompt learning,

    M. U. Khattak, H. Rasheed, M. Maaz, S. Khan, and F. S. Khan, “Maple: Multi-modal prompt learning,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023, pp. 19 113–19 122

  10. [18]

    Conditional prompt learning for vision-language models,

    K. Zhou, J. Yang, C. C. Loy, and Z. Liu, “Conditional prompt learning for vision-language models,” inProceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition, 2022, pp. 16 816– 16 825

  11. [19]

    Visual-language prompt tuning with knowledge-guided context optimization,

    H. Yao, R. Zhang, and C. Xu, “Visual-language prompt tuning with knowledge-guided context optimization,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023, pp. 6757–6767

  12. [20]

    Prompt-aligned gradient for prompt tuning,

    B. Zhu, Y . Niu, Y . Han, Y . Wu, and H. Zhang, “Prompt-aligned gradient for prompt tuning,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 15 659–15 669

  13. [21]

    Plot: Prompt learning with optimal transport for vision-language models,

    G. Chen, W. Yao, X. Song, X. Li, Y . Rao, and K. Zhang, “Plot: Prompt learning with optimal transport for vision-language models,” arXiv preprint arXiv:2210.01253, 2022

  14. [22]

    Self-regulating prompts: Foundational model adaptation without forgetting,

    M. U. Khattak, S. T. Wasim, M. Naseer, S. Khan, M.-H. Yang, and F. S. Khan, “Self-regulating prompts: Foundational model adaptation without forgetting,” inProceedings of the IEEE/CVF international conference on computer vision, 2023, pp. 15 190–15 200

  15. [23]

    Clip-adapter: Better vision-language models with feature adapters,

    P. Gao, S. Geng, R. Zhang, T. Ma, R. Fang, Y . Zhang, H. Li, and Y . Qiao, “Clip-adapter: Better vision-language models with feature adapters,” International Journal of Computer Vision, vol. 132, no. 2, pp. 581–595, 2024

  16. [24]

    Tip-adapter: Training-free adaption of clip for few-shot classification,

    R. Zhang, W. Zhang, R. Fang, P. Gao, K. Li, J. Dai, Y . Qiao, and H. Li, “Tip-adapter: Training-free adaption of clip for few-shot classification,” inEuropean conference on computer vision. Springer, 2022, pp. 493– 510

  17. [25]

    Task residual for tuning vision-language models,

    T. Yu, Z. Lu, X. Jin, Z. Chen, and X. Wang, “Task residual for tuning vision-language models,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 10 899–10 909

  18. [26]

    A closer look at the few-shot adaptation of large vision-language models,

    J. Silva-Rodriguez, S. Hajimiri, I. Ben Ayed, and J. Dolz, “A closer look at the few-shot adaptation of large vision-language models,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 23 681–23 690

  19. [27]

    V ocabulary-free few-shot learning for vision-language models,

    M. Zanella, C. Fuchs, I. Ben Ayed, and C. De Vleeschouwer, “V ocabulary-free few-shot learning for vision-language models,” inPro- ceedings of the Computer Vision and Pattern Recognition Conference, 2025, pp. 149–158

  20. [28]

    Low-rank few-shot adaptation of vision- language models,

    M. Zanella and I. Ben Ayed, “Low-rank few-shot adaptation of vision- language models,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 1593–1603

  21. [29]

    Aid: A benchmark data set for performance evaluation of aerial scene classification,

    G.-S. Xia, J. Hu, F. Hu, B. Shi, X. Bai, Y . Zhong, L. Zhang, and X. Lu, “Aid: A benchmark data set for performance evaluation of aerial scene classification,”IEEE Transactions on Geoscience and Remote Sensing, vol. 55, no. 7, pp. 3965–3981, 2017

  22. [30]

    Introducing eurosat: A novel dataset and deep learning benchmark for land use and land cover classification,

    P. Helber, B. Bischke, A. Dengel, and D. Borth, “Introducing eurosat: A novel dataset and deep learning benchmark for land use and land cover classification,” inIGARSS 2018-2018 IEEE international geoscience and remote sensing symposium. IEEE, 2018, pp. 204–207

  23. [31]

    Mlrsnet: A multi-label high spatial resolution remote sensing dataset for semantic scene understanding,

    X. Qi, P. Zhu, Y . Wang, L. Zhang, J. Peng, M. Wu, J. Chen, X. Zhao, N. Zang, and P. T. Mathiopoulos, “Mlrsnet: A multi-label high spatial resolution remote sensing dataset for semantic scene understanding,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 169, pp. 337...

  24. [32]

    Scene classification with recurrent attention of vhr remote sensing images,

    Q. Wang, S. Liu, J. Chanussot, and X. Li, “Scene classification with recurrent attention of vhr remote sensing images,”IEEE Transactions on Geoscience and Remote Sensing, vol. 57, no. 2, pp. 1155–1167, 2018

  25. [33]

    Patternnet: A benchmark dataset for performance evaluation of remote sensing image retrieval,

    W. Zhou, S. Newsam, C. Li, and Z. Shao, “Patternnet: A benchmark dataset for performance evaluation of remote sensing image retrieval,” ISPRS journal of photogrammetry and remote sensing, vol. 145, pp. 197–209, 2018

  26. [34]

    Remote sensing image scene classifi- cation: Benchmark and state of the art,

    G. Cheng, J. Han, and X. Lu, “Remote sensing image scene classifi- cation: Benchmark and state of the art,”Proceedings of the IEEE, vol. 105, no. 10, pp. 1865–1883, 2017

  27. [35]

    Feature significance-based multibag- of-visual-words model for remote sensing image scene classification,

    L. Zhao, P. Tang, and L. Huo, “Feature significance-based multibag- of-visual-words model for remote sensing image scene classification,” Journal of Applied Remote Sensing, vol. 10, no. 3, pp. 035 004–035 004, 2016

  28. [36]

    Rsi-cb: A large-scale remote sensing image classification benchmark using crowdsourced data,

    H. Li, X. Dou, C. Tao, Z. Wu, J. Chen, J. Peng, M. Deng, and L. Zhao, “Rsi-cb: A large-scale remote sensing image classification benchmark using crowdsourced data,”Sensors, vol. 20, no. 6, p. 1594, 2020

  29. [37]

    Structural high-resolution satellite image indexing,

    G.-S. Xia, W. Yang, J. Delon, Y . Gousseau, H. Sun, and H. Ma ˆıtre, “Structural high-resolution satellite image indexing,” inISPRS TC VII Symposium-100 Years ISPRS, vol. 38, 2010, pp. 298–303

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.