Pith. sign in

REVIEW 5 major objections 6 minor 43 references

RemoteTrimmer: Adaptive Structural Pruning for Remote Sensing Image Classification

T0 review · 5 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read RemoteTrimmer prunes remote-sensing classifiers to about 13 percent of their compute while keeping or beating original accuracy.

desk verdict A plausible pruning recipe with a coherent CAP idea, but the central loss in Eq. (4) is undefined as written, making the method unimplementable from the paper alone. read the letter →

arxiv 2412.12603 v2 pith:QXUWQZGC submitted 2024-12-17 cs.CV

Guangwenjie Zou , Liang Yao , Fan Liu , Chuanyi Zhang , Xin Li , Ning Chen , Shengxiang Xu , Jun Zhou This is my paper · ORCID
classification cs.CV
keywords structuralpruningremotesensingimageclassificationchannelattentionSqueeze-and-ExcitationadaptivemininglossmodelcompressionEuroSATUCMerced
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

RemoteTrimmer targets a specific failure: general-purpose pruning loses too much accuracy on remote sensing classifiers because top-down imagery compresses the differences between channel importances. To counter this, the paper prunes channels using an importance score that multiplies Squeeze-and-Excitation attention scores by batch-normalization scaling factors, amplifying the contrast between important and unimportant channels. It then fine-tunes the pruned model with an adaptive mining loss that up-weights hard samples. On EuroSAT and UCMerced, the paper reports that pruned ResNet18 and VGG16 match or exceed their unpruned accuracies while cutting multiply-accumulate operations to about 13 percent of the original on EuroSAT. If correct, this would make high-accuracy remote sensing classification practical on lightweight hardware.

What carries the argument

The carrier of the argument is the channel importance score $I_{i,k} = \bar{s}_{i,k} \cdot \gamma_{i,k}$, where $\bar{s}_{i,k}$ is the dataset-averaged Squeeze-and-Excitation attention score for channel $k$ at layer $i$ and $\gamma_{i,k}$ is that channel's batch-normalization scaling factor. Multiplying the two maps channel importance into an attention space in which important and unimportant channels separate more cleanly, and the $\alpha$-quantile rule decides which channels to remove. The second mechanism is the adaptive mining loss $L_{\mathrm{AM}}(y,\hat{y}) = r \cdot L_{\mathrm{CE}}(y,\hat{y}) + e^{|y-\hat{y}|/N + (|y-\hat{y}|/N)^2}$, whose exponential term is meant to concentrate fine-tuning on difficult samples. Together, these two mechanisms are what the ablation studies attribute the accuracy gains to.

What would settle it

Reproduce Table I with the AML term removed and cross-entropy fine-tuning under the exact same epochs, batch size, and learning-rate schedule; if pruned ResNet18 still reaches 0.922 on EuroSAT, then the reported accuracy gain is not caused by the adaptive mining loss. To test the loss itself, evaluate Eq. (4) on a one-hot EuroSAT label with the paper's own definitions; if $|y-\hat{y}|$ cannot be computed, the loss as written is not executable.

Watch

Extended reading notes

Core claim

The central claim is that a pruning method built for remote sensing imagery can make the pruned model as accurate as, and sometimes more accurate than, the model it came from. The recipe has two parts. First, Channel Attention Pruning (CAP) trains a Squeeze-and-Excitation block, averages its channel-attention scores over the dataset, and multiplies each score by the corresponding batch-normalization scaling factor $\gamma$ to form an importance score; channels below the $\alpha$-quantile are removed. Second, Adaptive Mining Loss (AML) adds an exponential error term to cross-entropy so that samples with large prediction errors receive more weight during fine-tuning. In the reported experiments, pruned ResNet18 reaches 0.922 accuracy on EuroSAT, up from the 0.870 unpruned baseline, and pruned VGG16 holds 0.957, with multiply-accumulate operations cut from 0.15 G to 0.02 G on EuroSAT for ResNet18. The paper reads this as evidence that remote sensing pruning needs both attention-amplified importance and hard-sample-aware fine-tuning.

Load-bearing premise

The loss in Eq. (4) must be a well-defined function that really up-weights hard samples, but the paper never says what $y$ and $\hat{y}$ are; for one-hot categorical labels $|y-\hat{y}|$ is meaningless, so the central fine-tuning mechanism rests on an unspecified quantity.

Editorial extensions

If this is right

  • Directly from the paper's results, pruning can improve accuracy: pruned ResNet18 on EuroSAT scores 0.922 versus 0.870 unpruned, a 5.2-point gain, while MACs drop from 0.15 G to 0.02 G.
  • Also directly, the gains reproduce qualitatively on a second architecture and dataset: VGG16 on EuroSAT stays flat at 0.957 after pruning, and ResNet18 on UCMerced edges from 0.849 to 0.853.
  • The ablation isolating CAP shows that the attention-mapped importance score accounts for most of the ResNet18 gain on EuroSAT, moving accuracy from 0.870 to 0.917.
  • The AML component outperforms Focal Loss and plain cross-entropy under the same fine-tuning budget, which supports the difficult-sample mining story.
  • At a 0.9 pruning rate, ResNet18 accuracy drops substantially, so the method's margin shrinks when the remaining model is very small.

Reading between the lines

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

  • The largest accuracy gain appears on EuroSAT, where the fine-tuning schedule is longer for ResNet18; the paper does not ablate the extra fine-tuning budget independently of AML, so part of the reported gain may be a training-schedule effect rather than the mining term.
  • Because CAP's importance score is simply an element-wise product of two scalars, it can be applied to any convolutional backbone; a natural untested extension is to object detection or segmentation models, where the same channel-importance compression should appear.
  • A reader would need explicit definitions of $y$ and $\hat{y}$ in Eq. (4) before trusting the AML term; for categorical labels the expression $|y-\hat{y}|$ is not defined, and without a soft-label interpretation the claimed mining behavior is not a closed form.
  • If the mechanism is right, then even simpler channel statistics such as average activation magnitudes might give a similar amplification when combined with the same fine-tuning loss; the paper only tests learned attention modules, so a cheap-statistics baseline would clarify what SENet specifically contributes.
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

5 major / 6 minor

Summary. The paper proposes RemoteTrimmer, a structural pruning method for remote sensing image classification, with two main components: a Channel Attention Pruning (CAP) strategy that combines SENet attention scores with Batch Normalization scaling factors to compute channel importance, and an Adaptive Mining Loss (AML) for fine-tuning the pruned model to emphasize difficult samples. Experiments on EuroSAT and UC Merced with ResNet18 and VGG16 report that the pruned and fine-tuned models match or exceed the accuracy of the unpruned baselines while substantially reducing MACs, and that the method outperforms several existing pruning baselines.

Significance. If the method is reproducible as claimed, it would be a practically relevant contribution to model compression for remote sensing image classification, and the paper's motivation—that remote sensing images have less distinct channel importance and contain more difficult samples—is reasonable. The authors provide a code repository, which is a strength. However, the manuscript in its current form has serious gaps: the central loss equation (4) is not well-defined for classification, the experimental setup omits data splits and training protocols, and the efficiency comparison in Table I lacks MACs/parameters for most methods. These issues must be resolved before the significance of the empirical results can be assessed.

major comments (5)
  1. [Section II-B, Eq. (4)] The Adaptive Mining Loss is not well-defined for classification. The variables y and ŷ are not specified: if y is a categorical label and ŷ is a predicted probability vector, the expression |y - ŷ| is undefined, and exponentiating a vector without a norm is not a scalar loss. The text also refers to 'LLI' and its derivative with respect to 'a certain pixel,' which is never reconciled with Eq. (4). Since Table II shows that AML alone improves ResNet18 accuracy from 0.870 to 0.917, this is the central fine-tuning component and must be stated precisely, including the intended soft-label or vector interpretation and a derivation of the adaptive mining behavior.
  2. [Section III-B] The experimental setup does not state the train/validation/test split for EuroSAT or UCM, nor does it describe the training schedule for the SENet-augmented models and the baseline models before pruning. Without this information, the reported accuracies cannot be reproduced, and the comparison with baseline pruning methods is not properly controlled.
  3. [Table I] MACs and Parameters are reported only for the baseline and the BN method; the L1-norm, FPGM, DepGraph, and Ours rows have empty cells for these metrics on both datasets. Since the paper's core claim is that pruning reduces computational cost, the table should report MACs and parameters for all compared methods to verify that the claimed accuracy gains are not obtained at a higher computational cost.
  4. [Section III-D5] The balance factor r is selected by maximizing EuroSAT accuracy (r=0.4 gives 92.2%) and then the same EuroSAT results are reported in Table I with this setting. This constitutes parameter selection on the evaluation benchmark, which can inflate the reported performance. The authors should either use a validation split for tuning r or present the sensitivity analysis without claiming the best r as the final setting.
  5. [Introduction, Contributions] The contribution statement claims 'this is the first pruning method for remote sensing image classification models,' but references [20]–[22] describe pruning methods for remote sensing scene classification. The novelty claim should be revised to acknowledge these prior works and clarify the specific difference of the proposed approach.
minor comments (6)
  1. [Fig. 2 caption] The caption mentions 'lateral inhibition loss function,' but the text uses 'Adaptive Mining Loss'; please unify the terminology.
  2. [Section II-B, Eq. (4)] The line defining variables says 'θ represents the parameters of the classification model,' but θ does not appear in Eq. (4); either remove it or include it in the loss expression.
  3. [Section III-D4] The text says 'as illustrated in Tab. IV' for the pruning-rate ablation, but Table IV is the balance-factor analysis; the pruning-rate results are actually shown in Fig. 3(b).
  4. [Section III-D1] The text refers to 'CAP and LIL components,' but LIL is not defined; it should be 'AML.'
  5. [Eq. (3)] The term 'Quantile α(I(C))' is ambiguous; please define it explicitly as the α-th quantile of the importance scores in layer C.
  6. [Tables I and II] The VGG16 baseline accuracy on EuroSAT is 0.957 in Table I but 0.954 in Table II; please clarify which value is correct and whether the tables use identical training settings.

Circularity Check

1 steps flagged · score 4.0 of 10

EuroSAT results are selected, not independently predicted: the balance factor r is tuned on EuroSAT and the same EuroSAT accuracy is then reported as the method's headline result.

  1. fitted input called prediction [Section III-D5 'Parameter Analysis of Balance Factors' and Table I]
    "To ascertain the optimal balance factor value for r in the adaptive mining loss function, we conducted parameter analysis on EuroSAT dataset with ResNet18 and VGG16 as the base model. The results are illustrated in Tab IV. When r is set to 0.4, the model achieved the optimal accuracy of 92.2%. Therefore, we ultimately adopted this set of values in our method."

    The only free hyperparameter of the proposed loss is selected by maximizing Accuracy on EuroSAT (Table IV: r=0.4 gives 92.2% for ResNet18 and 95.7% for VGG16). Those exact EuroSAT accuracies then reappear in Table I's 'Ours' rows (0.922 and 0.957) as evidence of state-of-the-art performance. The headline EuroSAT result is not an independent evaluation of a fixed method; it is the optimum of a one-dimensional search on the same benchmark. No separate validation split is described, so the reported accuracy is the selection criterion itself. This is benchmark-level fitting rather than a derivation that reduces one equation to another, but it makes the EuroSAT portion of the central claim statistically forced.

full rationale

The pruning pipeline itself is not circular: CAP derives channel importance from SENet attention scores multiplied by BN scaling factors, which is standard target-trained importance estimation, and the pruned model is then fine-tuned and evaluated in the usual supervised way. No load-bearing self-citation chain appears; the cited remote-sensing motivation and attention modules are external or standard. Equation (4) is under-specified for classification because y and y-hat are never defined for categorical labels, and the text switches to 'LLI' and pixel-wise derivative language, but that is an implementability/reproducibility defect, not a circular reduction. The only circularity-adjacent issue is the selection of r on EuroSAT and the reuse of that same EuroSAT accuracy as the headline result; UCM results, where r is transferred, retain some independent content. I therefore score the paper 4 rather than higher.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central claim rests on two tunable numbers (r and alpha), on a broad empirical assumption about remote sensing channels, and on a loss formula whose variables are never defined. No new physical or conceptual entities are introduced.

free parameters (2)
  • balance factor r in AML = 0.4
    Chosen based on EuroSAT accuracy in Table IV; the final reported EuroSAT results use this value, so the evaluation benchmark partly guides the constant.
  • pruning rate alpha = 0.7
    Fixed at 0.7 for all experiments except the pruning-rate ablation; this choice is not derived from first principles.
assumptions (3)
  • domain assumption Remote sensing models have compressed inter-channel importance differences that attention mapping can usefully amplify.
    Motivates CAP; supported only by citations [15] and [18], not by a quantitative test in this paper.
  • domain assumption After pruning, focusing fine-tuning on difficult samples is the right way to recover accuracy.
    Motivates AML; the paper does not provide a derivation, only an empirical loss comparison on one dataset.
  • ad hoc to paper Eq. (4) is a well-defined loss for classification, with y and ŷ quantities whose absolute difference is meaningful.
    The paper never defines y and ŷ for categorical labels, so the loss cannot be evaluated as written.

how reviews work

0 comments
Cite this review

Pith. "Pith review of RemoteTrimmer: Adaptive Structural Pruning for Remote Sensing Image Classification." pith.science (2026). https://pith.science/paper/QXUWQZGC

@misc{pith2026241212603,
  author       = {Pith},
  title        = {Pith review of: RemoteTrimmer: Adaptive Structural Pruning for Remote Sensing Image Classification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QXUWQZGC}},
  note         = {Machine review of arXiv:2412.12603}
}
read the original abstract

Since high resolution remote sensing image classification often requires a relatively high computation complexity, lightweight models tend to be practical and efficient. Model pruning is an effective method for model compression. However, existing methods rarely take into account the specificity of remote sensing images, resulting in significant accuracy loss after pruning. To this end, we propose an effective structural pruning approach for remote sensing image classification. Specifically, a pruning strategy that amplifies the differences in channel importance of the model is introduced. Then an adaptive mining loss function is designed for the fine-tuning process of the pruned model. Finally, we conducted experiments on two remote sensing classification datasets. The experimental results demonstrate that our method achieves minimal accuracy loss after compressing remote sensing classification models, achieving state-of-the-art (SoTA) performance.

Figures

Figures reproduced from arXiv: 2412.12603 by the authors.

Figure 1
Figure 1. Motivations of our proposed method. (a) Comparison of general (ImageNet) and remote sensing (UCM) images. Remote sensing images have top-down views from different heights, resulting in greater scale variations of the objects. This situation usually leads to the narrowing of differences between similar object features. (b) Remote sensing images often suffer from unclear images with atmospheric noise pollution, which … view at source ↗
Figure 2
Figure 2. Overview of our RemoteTrimmer. In the pruning phase, we leverage [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Ablation studies on different attention mechanisms (a) and pruning [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 25 canonical work pages

  1. [20]

    Pruning convolutional neural networks with an attention mechanism for remote sensing image classi- fication,

    S. Zhang, G. Wu, J. Gu, and J. Han, “Pruning convolutional neural networks with an attention mechanism for remote sensing image classi- fication,” Electronics, vol. 9, no. 8, p. 1209, 2020

    work page 2020

  2. [22]

    Data customization-based multiobjective optimization pruning framework for remote sensing scene classification,

    Z. Hu, M. Gong, Y . Lu, J. Li, Y . Zhao, and M. Zhang, “Data customization-based multiobjective optimization pruning framework for remote sensing scene classification,” IEEE Transactions on Geoscience and Remote Sensing , 2023

    work page 2023

  3. [1]

    The national entiron- mental and geological information system for remote sensing survey and monitoring,

    Y . Yan, Z. He, G. Liu, Y . Wang, and C. Han, “The national entiron- mental and geological information system for remote sensing survey and monitoring,” in 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) . IEEE, 2015, pp. 4700–4703

    work page 2015

  4. [2]

    Gross and net primary production in the global ocean: An ocean color remote sensing perspective,

    T. K. Westberry, G. M. Silsbe, and M. J. Behrenfeld, “Gross and net primary production in the global ocean: An ocean color remote sensing perspective,” Earth-Science Reviews, vol. 237, p. 104322, 2023

    work page 2023

  5. [3]

    Remote sensing of soil degradation: Progress and perspective,

    J. Wang, J. Zhen, W. Hu, S. Chen, I. Lizaga, M. Zeraatpisheh, and X. Yang, “Remote sensing of soil degradation: Progress and perspective,” International Soil and Water Conservation Research , vol. 11, no. 3, pp. 429–454, 2023

    work page 2023

  6. [4]

    A survey of machine learning and deep learning in remote sensing of geological environment: Challenges, advances, and opportunities,

    W. Han, X. Zhang, Y . Wang, L. Wang, X. Huang, J. Li, S. Wang, W. Chen, X. Li, R. Feng et al. , “A survey of machine learning and deep learning in remote sensing of geological environment: Challenges, advances, and opportunities,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 202, pp. 87–113, 2023

    work page 2023

  7. [5]

    Geological survey of urban roadbeds utilizing rapid detection system based on transient electromagnetic method,

    Y . An, Y . Wang, C. Liu, X. Zhang, S. Liu, and L. Du, “Geological survey of urban roadbeds utilizing rapid detection system based on transient electromagnetic method,” IEEE Transactions on Geoscience and Remote Sensing, 2024

    work page 2024

  8. [6]

    High-resolution remote sensing image scene understanding: A review,

    Q. Zhu, X. Sun, Y . Zhong, and L. Zhang, “High-resolution remote sensing image scene understanding: A review,” in IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium. IEEE, 2019, pp. 3061–3064

    work page 2019

Show all 43 references
  1. [7]

    Remote sensing image classification: A comprehensive review and applications,

    M. Mehmood, A. Shahzad, B. Zafar, A. Shabbir, and N. Ali, “Remote sensing image classification: A comprehensive review and applications,” Mathematical Problems in Engineering , vol. 2022, no. 1, p. 5880959, 2022

    2022

  2. [8]

    A review of ensemble learning algo- rithms used in remote sensing applications,

    Y . Zhang, J. Liu, and W. Shen, “A review of ensemble learning algo- rithms used in remote sensing applications,” Applied Sciences, vol. 12, no. 17, p. 8654, 2022

    2022

  3. [9]

    Transfer learning in environmental remote sensing,

    Y . Ma, S. Chen, S. Ermon, and D. B. Lobell, “Transfer learning in environmental remote sensing,” Remote Sensing of Environment , vol. 301, p. 113924, 2024

    2024

  4. [10]

    Learning both weights and con- nections for efficient neural network,

    S. Han, J. Pool, J. Tran, and W. Dally, “Learning both weights and con- nections for efficient neural network,” Advances in neural information processing systems, vol. 28, 2015

    2015

  5. [11]

    Distilling the knowledge in a neural network,

    G. Hinton, O. Vinyals, and J. Dean, “Distilling the knowledge in a neural network,” arXiv preprint arXiv:1503.02531 , 2015

    2015 arXiv

  6. [12]

    Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding,

    S. Han, H. Mao, and W. J. Dally, “Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding,” arXiv preprint arXiv:1510.00149 , 2015

    2015 arXiv

  7. [13]

    Pruning filters for efficient convnets,

    D. Filters’Importance, “Pruning filters for efficient convnets,” 2016

    2016

  8. [14]

    Depgraph: Towards any structural pruning,

    G. Fang, X. Ma, M. Song, M. B. Mi, and X. Wang, “Depgraph: Towards any structural pruning,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2023, pp. 16 091–16 101

    2023

  9. [15]

    Object detection in optical remote sensing images: A survey and a new benchmark,

    K. Li, G. Wan, G. Cheng, L. Meng, and J. Han, “Object detection in optical remote sensing images: A survey and a new benchmark,” ISPRS journal of photogrammetry and remote sensing , vol. 159, pp. 296–307, 2020

    2020

  10. [16]

    Convolutional neural networks for large-scale remote-sensing image classification,

    E. Maggiori, Y . Tarabalka, G. Charpiat, and P. Alliez, “Convolutional neural networks for large-scale remote-sensing image classification,” IEEE Transactions on geoscience and remote sensing , vol. 55, no. 2, pp. 645–657, 2016

    2016

  11. [17]

    A critical review of high and very high-resolution remote sensing approaches for detecting and mapping slums: Trends, challenges and emerging opportunities,

    R. Mahabir, A. Croitoru, A. T. Crooks, P. Agouris, and A. Stefanidis, “A critical review of high and very high-resolution remote sensing approaches for detecting and mapping slums: Trends, challenges and emerging opportunities,” Urban Science, vol. 2, no. 1, p. 8, 2018

    2018

  12. [18]

    Improved remote sensing image classification based on multi- scale feature fusion,

    C. Zhang, Y . Chen, X. Yang, S. Gao, F. Li, A. Kong, D. Zu, and L. Sun, “Improved remote sensing image classification based on multi- scale feature fusion,” Remote Sensing, vol. 12, no. 2, p. 213, 2020

    2020

  13. [19]

    Multi-scale dense networks for hyper- spectral remote sensing image classification,

    C. Zhang, G. Li, and S. Du, “Multi-scale dense networks for hyper- spectral remote sensing image classification,” IEEE Transactions on Geoscience and Remote Sensing , vol. 57, no. 11, pp. 9201–9222, 2019

    2019

  14. [21]

    Nas-based cnn channel pruning for remote sensing scene classification,

    X. Wei, N. Zhang, W. Liu, and H. Chen, “Nas-based cnn channel pruning for remote sensing scene classification,” IEEE Geoscience and Remote Sensing Letters, vol. 19, pp. 1–5, 2022

    2022

  15. [23]

    A light-weight deep learning model for remote sensing image classification,

    L. Pham, C. Le, D. Ngo, A. Nguyen, J. Lampert, A. Schindler, and I. McLoughlin, “A light-weight deep learning model for remote sensing image classification,” in 2023 International Symposium on Image and Signal Processing and Analysis (ISPA) . IEEE, 2023, pp. 1–6

    2023

  16. [24]

    Rscnet: An efficient remote sensing scene classification model based on lightweight convolution neural networks,

    Z. Chen, J. Yang, Z. Feng, and L. Chen, “Rscnet: An efficient remote sensing scene classification model based on lightweight convolution neural networks,” Electronics, vol. 11, no. 22, p. 3727, 2022

    2022

  17. [25]

    Cloud and cloud shadow detection for optical satellite imagery: Features, al- gorithms, validation, and prospects,

    Z. Li, H. Shen, Q. Weng, Y . Zhang, P. Dou, and L. Zhang, “Cloud and cloud shadow detection for optical satellite imagery: Features, al- gorithms, validation, and prospects,” ISPRS Journal of Photogrammetry and Remote Sensing , vol. 188, pp. 89–108, 2022

    2022

  18. [26]

    Remote sensing image restoration using various techniques: A review,

    N. Gulat and A. Kaushik, “Remote sensing image restoration using various techniques: A review,” International Journal of Scientific & Engineering Research, vol. 3, no. 1, p. 6, 2012

    2012

  19. [27]

    Atmospheric correction algorithms for hyperspectral remote sensing data of land and ocean,

    B.-C. Gao, M. J. Montes, C. O. Davis, and A. F. Goetz, “Atmospheric correction algorithms for hyperspectral remote sensing data of land and ocean,” Remote sensing of environment , vol. 113, pp. S17–S24, 2009

    2009

  20. [28]

    Comprehensive investigation of the atmospheric modulation transfer function (mtf) for satellite imaging pay- loads: considering turbulence and aerosol effects over tehran,

    H. Hosseini and M. Khoshsima, “Comprehensive investigation of the atmospheric modulation transfer function (mtf) for satellite imaging pay- loads: considering turbulence and aerosol effects over tehran,” Physica Scripta, vol. 99, no. 7, p. 075044, 2024

    2024

  21. [29]

    Visual attention methods in deep learning: An in-depth survey,

    M. Hassanin, S. Anwar, I. Radwan, F. S. Khan, and A. Mian, “Visual attention methods in deep learning: An in-depth survey,” Information Fusion, vol. 108, p. 102417, 2024

    2024

  22. [30]

    Squeeze-and-excitation networks,

    J. Hu, L. Shen, and G. Sun, “Squeeze-and-excitation networks,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 7132–7141

    2018

  23. [31]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 770–778

    2016

  24. [32]

    Deep sparse rectifier neural networks,

    X. Glorot, A. Bordes, and Y . Bengio, “Deep sparse rectifier neural networks,” in Proceedings of the fourteenth international conference on artificial intelligence and statistics . JMLR Workshop and Conference Proceedings, 2011, pp. 315–323

    2011

  25. [33]

    Learning efficient convolutional networks through network slimming,

    Z. Liu, J. Li, Z. Shen, G. Huang, S. Yan, and C. Zhang, “Learning efficient convolutional networks through network slimming,” inProceed- ings of the IEEE international conference on computer vision , 2017, pp. 2736–2744

    2017

  26. [34]

    Pruning filters for efficient convnets,

    H. Li, A. Kadav, I. Durdanovic, H. Samet, and H. P. Graf, “Pruning filters for efficient convnets,” arXiv preprint arXiv:1608.08710 , 2016

    2016 arXiv

  27. [35]

    Filter pruning via geometric median for deep convolutional neural networks acceleration,

    Y . He, P. Liu, Z. Wang, Z. Hu, and Y . Yang, “Filter pruning via geometric median for deep convolutional neural networks acceleration,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 4340–4349

    2019

  28. [36]

    Focal loss for dense object detection,

    T.-Y . Lin, P. Goyal, R. Girshick, K. He, and P. Doll ´ar, “Focal loss for dense object detection,” in Proceedings of the IEEE international conference on computer vision , 2017, pp. 2980–2988

    2017

  29. [37]

    Eurosat: A novel dataset and deep learning benchmark for land use and land cover classi- fication,

    P. Helber, B. Bischke, A. Dengel, and D. Borth, “Eurosat: A novel dataset and deep learning benchmark for land use and land cover classi- fication,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2019

    2019

  30. [38]

    Deep semantic understanding of high resolution remote sensing image,

    B. Qu, X. Li, D. Tao, and X. Lu, “Deep semantic understanding of high resolution remote sensing image,” in 2016 International conference on computer, information and telecommunication systems (Cits) . IEEE, 2016, pp. 1–5

    2016

  31. [39]

    Towards performance-maximizing neural network pruning via global channel attention,

    Y . Wang, S. Guo, J. Guo, J. Zhang, W. Zhang, C. Yan, and Y . Zhang, “Towards performance-maximizing neural network pruning via global channel attention,” Neural Networks, vol. 171, pp. 104–113, 2024

    2024

  32. [40]

    Very deep convolutional networks for large-scale image recognition,

    K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556 , 2014

    2014 arXiv

  33. [41]

    Sgd: Saccharomyces genome database,

    J. M. Cherry, C. Adler, C. Ball, S. A. Chervitz, S. S. Dwight, E. T. Hes- ter, Y . Jia, G. Juvik, T. Roe, M. Schroeder et al., “Sgd: Saccharomyces genome database,” Nucleic acids research , vol. 26, no. 1, pp. 73–79, 1998

    1998

  34. [42]

    Cbam: Convolutional block attention module,

    S. Woo, J. Park, J.-Y . Lee, and I. S. Kweon, “Cbam: Convolutional block attention module,” in Proceedings of the European conference on computer vision (ECCV) , 2018, pp. 3–19

    2018

  35. [43]

    Bam: Bottleneck attention module,

    J. Park, “Bam: Bottleneck attention module,” arXiv preprint arXiv:1807.06514, 2018

    2018 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.