Hyperspectral Images Efficient Spatial and Spectral non-Linear Model with Bidirectional Feature Learning

Chenhong Sui; Jing Wang; Judy X Yang; Jun Zhou; Zekun Long

REVIEW 5 major objections 7 minor 32 references

Hyperspectral Images Efficient Spatial and Spectral non-Linear Model with Bidirectional Feature Learning

T0 review · 5 major / 7 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a lightweight bidirectional CNN, which scans spectral bands forward and backward and fuses the results with a spatial convolution block, achieves higher hyperspectral image classification accuracy than transformer…

desk verdict The paper's architecture is a known RNN/CNN hybrid and its headline accuracy claims are undermined by internal contradictions and an unfair comparison protocol. read the letter →

arxiv 2412.00283 v2 pith:WLW7B3OC submitted 2024-11-29 cs.CV

Judy X Yang , Jing Wang , Zekun Long , Chenhong Sui , Jun Zhou This is my paper · ORCID

classification cs.CV

keywords HyperspectralimageclassificationBidirectionalnetworksSpectral-spatialfeaturelearningComputingefficiencyState-spacemodelsTransformersRemotesensingPatchsize

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

Hyperspectral images carry hundreds of narrow spectral bands, making classification accurate but computationally heavy. The paper proposes the Spectral-Spatial non-Linear Model (SS non-Linear Model), which processes the spectral dimension twice—once forward and once backward with 1D convolutions—and then adds a 2D spatial block before classification. The authors claim this bidirectional CNN reaches higher overall accuracy than transformer-based competitors on Houston 2013 (97.60% vs 96.35% for HiT), Indian Pines (93.32% vs 92.53%), and Pavia University (98.07% vs 92.00%), with FLOPs of 0.05 GB versus 2.33 GB for HiT. If the comparison holds, accurate hyperspectral classification could run on resource-constrained hardware without sacrificing accuracy.

What carries the argument

The load-bearing mechanism is the bidirectional spectral processing block. The input patch is projected into two hidden states; one scans the spectral bands in the natural order and the other scans the reversed band order, each step a 1D convolution followed by a tanh non-linearity that is modulated by learned matrices A and B (Eqs. 8-9). The forward and backward streams are averaged (Eq. 10), giving a spectral feature vector that is then concatenated with the output of a 2D spatial convolution block. This design replaces self-attention with two cheap 1D convolutional passes, which is what the paper credits for the large FLOPs reduction relative to transformers.

What would settle it

Re-run the three benchmark comparisons with all models at the same patch size (say 15) and also with each model's individually tuned patch size, repeating each run over at least five random seeds and reporting mean and standard deviation of overall accuracy. If the SS non-Linear Model's accuracy no longer exceeds HiT by more than the run-to-run noise, or if its FLOPs advantage shrinks when measured at its chosen patch size on the same hardware, then the paper's central claim of superior accuracy with efficiency would fail.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that a deliberately lightweight bidirectional spectral-scanning neural network can beat transformer-based models at hyperspectral image classification. The SS non-Linear Model projects each spatial-spectral patch into two streams, scans the spectral bands forward under transformation A and backward under transformation B, combines the two hidden states, and passes the result through a spatial convolution block before a fully connected classifier. The reported experiments on three standard datasets give overall accuracies of 97.60% on Houston 2013 (patch 5), 93.32% on Indian Pines (patch 5), and 98.07% on Pavia University (patch 15), outperforming the HiT transformer baseline on each dataset and also beating several CNN, RNN, and vision-transformer variants. The paper attributes this to bidirectional spectral context plus spatial fusion, and reports the lowest FLOPs (0.05 GB) of all compared models.

Load-bearing premise

The accuracy comparison assumes every model is measured under fairly matched conditions, but the paper tunes the proposed model's patch size per dataset while fixing all baselines at patch 15, and reports single-run accuracy without error bars.

Editorial extensions

If this is right

On the three tested datasets, hyperspectral classification can reach or exceed transformer accuracy with patch sizes of 5 or smaller, meaning memory use and inference time scale down substantially.
The bidirectional spectral scan appears to capture information that unidirectional CNNs and attention-based models miss, since removing either direction costs 2-3% overall accuracy in the ablation study.
The reported FLOPs of 0.05 GB versus 2.33 GB for HiT suggest the architecture could be deployed on drones, edge devices, or real-time monitoring systems where transformer models are impractical.
The model's compact size and fast inference make it a candidate for large-scale or real-time hyperspectral analysis tasks such as precision agriculture and urban monitoring.

Reading between the lines

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

If the patch-size protocol were equalized—baselines also tuned per dataset or the proposed model fixed at patch 15—the reported accuracy margins over HiT might shrink; the claimed advantage is therefore tied to the tuning protocol, not solely to the architecture.
The same bidirectional spectral-scanning design could be tested on other high-dimensional signals with a dominant ordering axis, such as multispectral video frames or spectrograms, as a low-cost alternative to attention in those domains.
A stronger efficiency comparison would measure FLOPs and runtime at each model's own operating patch size and input resolution rather than a fixed 1×15×15×200 input; that comparison could change the relative efficiency ranking.

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

Colleague, quick read on arXiv:2412.00283. The punchline: the claimed accuracy advantage over HiT is not supported by the paper's own numbers. The text and tables disagree on the key results, and the comparison protocol gives the proposed model an advantage by tuning its patch size on the test set while fixing baselines at patch 15. The architecture itself is a straightforward bidirectional RNN with 1D convolutions plus a 2D CNN branch; the Mamba/SSM framing in the introduction is not realized in the equations.

What's actually new: the specific combination and the empirical results, if they hold. The paper does run a patch-size sensitivity study and an ablation over forward/backward/spatial branches, which is the right kind of analysis. The idea of a lightweight bidirectional spectral branch is reasonable and could be efficient.

The soft spots are serious and load-bearing. First, the paper contradicts itself. On Houston, Section IV.C says Method 1 trails HiT, but Table IV shows it ahead (97.57 vs 96.35). On Indian Pines, the text reports 92.41/92.18 but Table V shows 93.32/92.53. On Pavia, the text reports 98.14/96.35 but Table VI shows 97.04/98.07. These aren't typos; they reverse the conclusions. Second, the comparison is unfair: baselines use patch 15, while the proposed model's patch size is chosen per dataset from a test-set sweep (P5 for Houston, P11 for Indian Pines, P13 for Pavia). That alone can account for the claimed gains. Third, there are no seeds or error bars, so a 1.25-point margin is noise. No code is linked despite the abstract promising a repository. Fourth, the efficiency table uses 200 bands while Houston has 144, and the claimed 40% FLOP reduction is contradicted by the paper's own complexity formulas. There's also an undefined delta_expanded in Equations (8)-(9), and the ablation is described in the text as patch size 3 but the table labels it patch size 5.

Who is this for? A reader who wants a quick look at a simple efficient HSI classifier might skim the architecture, but they should treat every number as unverified. This paper does not deserve a serious referee in its current form. The correct move is to desk reject, with the authors advised to rerun with a controlled protocol, multiple seeds, and released code. The underlying architecture might be fine, but the paper as written cannot be evaluated.

Referee Report

5 major / 7 minor

Summary. The paper proposes a Spectral-Spatial non-Linear Model for hyperspectral image classification, combining bidirectional spectral processing via 1D convolutions and transformation matrices A and B with a spatial CNN block and a classifier. The central claim, stated in the Abstract and Section IV.C, is that this model yields superior classification accuracy compared with transformer and CNN baselines while being computationally efficient, validated on Houston 2013, Indian Pines, and Pavia University. The manuscript reports two variants of the model, ablation studies, patch-size sensitivity analyses, and an efficiency comparison. However, the headline numbers in the text contradict the tables on all three datasets, the comparison protocol is not controlled, the efficiency analysis is internally inconsistent, and no code, seeds, or error bars are provided. As a result, the paper's central claim is not established by the evidence presented.

Significance. If the claimed results were reliable, the paper would offer a useful contribution: a low-compute bidirectional CNN-style architecture that beats transformer baselines such as HiT at smaller patch sizes would be attractive for resource-constrained hyperspectral analysis. The architectural idea of combining bidirectional spectral state transitions with a spatial CNN block is plausible and worth exploring. However, the paper does not ship machine-checked proofs, reproducible code, or multi-seed evaluations, and the empirical tables that constitute the only evidence for the central claim are self-contradictory. The efficiency argument is also contradicted by the paper's own complexity formulas and by Table XI. I therefore see the current significance claim as unsupported by the manuscript in its present form.

major comments (5)

[Section IV.C, Tables IV-VI] The headline numbers in the text contradict the tables on all three datasets. In Houston, the text says Method 1 'closely trails' HiT and Method 2 surpasses HiT by about 0.5%, but Table IV shows OurMethod1 at 97.57% versus HiT at 96.35% (Method 1 is ahead by 1.22 points) and OurMethod2 at 97.60%. In Indian Pines, the text reports OAs of 92.41% (Method 2) and 92.18% (Method 1), whereas Table V reports 93.32% and 92.53%. In Pavia, the text reports 98.14% (Method 2) and 96.35% (Method 1), whereas Table VI reports 98.07% and 97.04%. These internal contradictions make it impossible to determine which numbers support the abstract's claim of superior accuracy.
[Section IV.B.2 and IV.D.1] The comparison protocol is not controlled: all baselines are fixed at patch size 15, while the proposed Method 2 is reported at patch size 3 in some parts of the Section IV.C text, patch size 5 in the captions of Tables IV-VI, and the Section IV.D.1 discussion selects patch sizes per dataset from test-set OA sweeps (e.g., 'P5 Houston, P11 Indian Pines, P13 Pavia' in the text, though the tables themselves show multiple peaks). This means the proposed model's patch size is effectively tuned on test accuracy while baselines are not. A valid comparison would require the same patch-size selection rule for all methods, or a proper held-out validation set, and the text is additionally inconsistent about whether Method 2 uses patch size 3 or 5.
[Section III.C, Eqs. (1)-(10), Table XI] The efficiency analysis is internally inconsistent. The paper gives the SS non-Linear Model FLOPs as O(Batch.H.W * CH) and claims a roughly 40% reduction relative to CNNs for kernel size k>3; under the paper's own CNN formula O(Batch.H.W * k^2 * CH), the reduction is 1 - 1/k^2, which for k=3 is about 89%, not 40%. The parameter-count claim O(CH+HW) is also contradicted by the transformation matrices A,B in R^{N*N} used in Eqs. (8)-(9), which contribute O(CH^2) parameters when N=CH. Table XI shows the proposed model's training time (130.30 s) exceeding R-2D-CNN (31.6 s) and 2D-CNN (15.92 s), so the claim of 'maintaining computational efficiency' is not supported by the reported numbers.
[Section IV (general)] No repeated runs, seeds, or error bars are reported, and the promised GitHub repository (HSILinearModel) is not linked. The claimed margins over HiT are small (for example, 1.25 points on Houston 2013), and without run-to-run variance information these differences cannot be distinguished from noise. The absence of a linked code repository further prevents any independent check of the reported results, so the central empirical claim is not reproducible from the manuscript.
[Section III.B, Eqs. (8)-(10)] Core operations of the proposed method are left undefined. The 'delta-modulated non-linearity' is never defined; the symbol Delta_expanded is not introduced; the reduce operation in Eqs. (5) and (10) is unspecified; and the dimensions of the transformation matrices A and B in Eqs. (8)-(9) are inconsistent with the earlier definition A,B in R^{N*N} in Section III.A. Without these definitions, the architecture cannot be implemented or assessed for novelty.

minor comments (7)

[Section III.A] The heading contains a doubled word: 'SS non-Linear Model Model Preliminaries'.
[Section IV.C] The Houston text states that Method 2 uses a patch size of 3, while Table IV's caption and the reported OA of 97.60% correspond to patch size 5; the Pavia text similarly says patch size 3 while Table VI's caption says patch size 5.
[Section IV.D.1] The text says 'Method 2, which uses P3 across all datasets' but Tables IV-VI report OurMethod2 at patch size 5, and the preceding paragraph gives different optimal patch sizes per dataset.
[Section IV.C] The sentence 'Its performance is higher than 5% of the other competitors' is not a meaningful quantitative claim and should be rephrased.
[Section IV.D.2, Table X] The text says the ablation study used a patch size of 3, but Table X is captioned 'Patch Size 5'; the text also describes the OA drops for removing forward/backward pathways with numbers that do not match Table X.
[Section II.C] The related work says Mamba demonstrates 'non-linear scalability' and later 'linear computational scalability' in adjacent sentences; the intended property is presumably linear scalability.
[Abstract and Section IV] The GitHub repository is mentioned by name but no URL or repository identifier is given, making it impossible to locate.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's claims rest on empirical comparisons and explicitly specified model equations, not on a derivation that reduces to its inputs.

full rationale

The paper's central claims are empirical (abstract; Section IV.C): the SS non-Linear Model reports higher OA than HiT on Houston (97.60% vs 96.35%, Table IV), 93.32% on Indian Pines (Table V), and 98.07% on Pavia (Table VI) with lower FLOPs (Table XI). These are benchmark results, not analytical derivations from a fitted quantity. The architecture is defined by explicit equations (Eqs. 1-13), including trainable matrices A and B, 1D convolutions, and spatial convolutions; no equation is defined in terms of the accuracy numbers it is said to predict. The related-work self-citations ([10], [11] by Wang/Zhou/Yang) are used only as background for band selection, and band selection is not a component of the proposed model, so they are not load-bearing. The Mamba citation is used as inspiration, not as a theorem that forces the architecture. Concerns raised by the text itself—the per-dataset patch size chosen from test-set OA in Section IV.D.1, single-run metrics without error bars, the text/table inconsistencies in OA values, and the unverifiable '40% FLOP reduction'—are experimental-fairness and correctness risks, not circular reductions: no prediction is constructed from a fitted parameter or from the authors' prior results. I therefore find no circularity under the specified definitions.

Assumptions & free parameters 5 free parameters · 4 assumptions · 1 invented entities

The central claims rest on learned matrices A and B plus an undefined delta in the core pathway, per-dataset patch sizes chosen from the test-set sweep, a fairness assumption that fixing baselines at patch 15 is an even comparison, and asserted FLOPs formulas that are internally inconsistent with the 40% reduction claim. These are the quantities the reader would need in order to reproduce or audit the reported results.

free parameters (5)

Per-dataset patch size (Method 2) = Houston P5 (text also says P3); Indian Pines P5; Pavia P5 (text also says P3)
Chosen per dataset from the test-set accuracy sweep (Tables VII-IX); reported as the model's result while baselines are fixed at patch 15.
Transformation matrices A and B = Learned; size not specified
Central to Equations (1)-(4) and (8)-(9); their dimension, initialization, and regularization are not given.
Delta expansion term (delta_expanded) = Undefined
Used in Equations (8)-(9) for the 'delta-modulated non-linearity'; never defined, so the core pathway is underspecified.
Hidden sizes and kernel widths = Not reported
Channel counts, kernel sizes, and layer counts for the Bi-Networks and spatial blocks are omitted, so Table XI FLOPs cannot be reproduced.
Training hyperparameters = lr=5e-4, batch=32, 100 epochs
Used for the proposed model (Section IV.B.3); no equivalent settings are reported for the baselines.

assumptions (4)

domain assumption A bidirectional spectral recurrence with 1D convolutions captures long-range spectral dependencies as well as attention or SSM blocks.
This is the architectural premise of Sections III.A-B; the Mamba inspiration in Section II.C is invoked but no selective-state-space mechanism is implemented.
ad hoc to paper The complexity formulas O(CH+HW) parameters and O(BHW*CH) FLOPs describe the proposed model.
Stated in Section III.C without derivation; the 40% reduction claim contradicts these formulas (k=3 gives a 9x gap) and Table XI (0.05 vs 0.07 GB for 2D-CNN).
domain assumption A single run per configuration with no seeds is sufficient to report OA and Kappa as the model's performance.
All results in Tables IV-XI are single-run; no variance or repeat trials are reported.
domain assumption The train/test splits are the standard splits for the three datasets.
Section IV.A reports class counts (e.g., Indian Pines total 10366) that differ from standard published counts and have internal arithmetic errors (Alfalfa row 6+45 listed as 54; train rows sum to 1101, not the stated 1061).

invented entities (1)

Delta-modulated non-linearity
purpose: Modulates the forward and backward spectral hidden states in Equations (8)-(9).
The term delta_expanded is introduced as a core operation but never defined, measured, or given a falsifiable consequence, so it cannot be implemented or independently verified.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Hyperspectral Images Efficient Spatial and Spectral non-Linear Model with Bidirectional Feature Learning." pith.science (2026). https://pith.science/paper/WLW7B3OC

@misc{pith2026241200283,
  author       = {Pith},
  title        = {Pith review of: Hyperspectral Images Efficient Spatial and Spectral non-Linear Model with Bidirectional Feature Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WLW7B3OC}},
  note         = {Machine review of arXiv:2412.00283}
}

read the original abstract

Classifying hyperspectral images (HSIs) is a complex task in remote sensing due to the high-dimensional nature and volume of data involved. To address these challenges, we propose the Spectral-Spatial non-Linear Model, a novel framework that significantly reduces data volume while enhancing classification accuracy. Our model employs a bidirectional reversed convolutional neural network (CNN) to efficiently extract spectral features, complemented by a specialized block for spatial feature analysis. This hybrid approach leverages the operational efficiency of CNNs and incorporates dynamic feature extraction inspired by attention mechanisms, optimizing performance without the high computational demands typically associated with transformer-based models. The SS non-Linear Model is designed to process hyperspectral data bidirectionally, achieving notable classification and efficiency improvements by fusing spectral and spatial features effectively. This approach yields superior classification accuracy compared to existing benchmarks while maintaining computational efficiency, making it suitable for resource-constrained environments. We validate the SS non-Linear Model on three widely recognized datasets, Houston 2013, Indian Pines, and Pavia University, demonstrating its ability to outperform current state-of-the-art models in HSI classification and efficiency. This work highlights the innovative methodology of the SS non-Linear Model and its practical benefits for remote sensing applications, where both data efficiency and classification accuracy are critical. For further details, please refer to our code repository on GitHub: HSILinearModel.

Figures

Figures reproduced from arXiv: 2412.00283 by the authors.

**Figure 2.** classification maps for the Houston 2013 dataset. Ground-Truth map, 9 comparative methods and Our SS non-Linear [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗

**Figure 3.** Visualization and classification maps for the Indian Pines dataset. Ground-Truth map, 9 comparative methods and Our [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Visualization and classification maps for the University Pavia dataset. Ground-Truth map, 9 comparative methods and [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]

**Figure 5.** Figure 5: OA Performance Comparison of Different patch size [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 22 canonical work pages

[1]

Hyperspectral image super-resolution meets deep learning: A survey and perspective,

X. Wang, Q. Hu, Y . Cheng, and J. Ma, “Hyperspectral image super-resolution meets deep learning: A survey and perspective,” IEEE/CAA Journal of Automatica Sinica , vol. 10, no. 8, pp. 1668–1691, 2023

work page 2023
[2]

Bands sensitive convolutional network for hyperspectral image classification,

L. Ran, Y . Zhang, W. Wei, and T. Yang, “Bands sensitive convolutional network for hyperspectral image classification,” in Proceedings of the International Conference on Internet Multimedia Computing and Service , 2016, pp. 268–272

work page 2016
[3]

Ikeuchi, Computer vision: A reference guide

K. Ikeuchi, Computer vision: A reference guide . Springer, 2021

work page 2021
[4]

The role of hyperspectral imaging: A literature review,

M. Mateen, J. Wen, M. A. Akbar et al., “The role of hyperspectral imaging: A literature review,” International Journal of Advanced Computer Science and Applications , vol. 9, no. 8, 2018

work page 2018
[5]

A review on the combination of deep learning techniques with proximal hyperspectral images in agriculture,

J. G. A. Barbedo, “A review on the combination of deep learning techniques with proximal hyperspectral images in agriculture,” Computers and Electronics in Agriculture , vol. 210, p. 107920, 2023

work page 2023
[6]

Automatic apple recognition based on the fusion of color and 3D feature for robotic fruit picking,

Y . Tao and J. Zhou, “Automatic apple recognition based on the fusion of color and 3D feature for robotic fruit picking,” Computers and Electronics in Agriculture, vol. 142, pp. 388–396, 2017

work page 2017
[7]

K. P. Vadrevu, T. Le Toan, S. S. Ray, and C. O. Justice, Remote Sensing of Agriculture and Land Cover/Land Use Changes in South and Southeast Asian Countries . Springer, 2022

work page 2022
[8]

Land use and land cover classification with hyperspectral data: A comprehensive review of methods, challenges and future directions,

M. A. Moharram and D. M. Sundaram, “Land use and land cover classification with hyperspectral data: A comprehensive review of methods, challenges and future directions,” Neurocomputing, 2023

work page 2023

Show all 32 references

[9]

Bs-nets: An end-to-end framework for band selection of hyperspectral image,

Y . Cai, X. Liu, and Z. Cai, “Bs-nets: An end-to-end framework for band selection of hyperspectral image,” IEEE Transactions on Geoscience and Remote Sensing , vol. 58, no. 3, pp. 1969–1984, 2019

1969
[10]

Attend in bands: Hyperspectral band weighting and selection for image classification,

J. Wang, J. Zhou, and W. Huang, “Attend in bands: Hyperspectral band weighting and selection for image classification,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , vol. 12, no. 12, pp. 4712–4727, 2019

2019
[11]

Lidar-guided cross-attention fusion for hyperspectral band selection and image classification,

J. X. Yang, J. Zhou, J. Wang, H. Tian, and A. W. C. Liew, “Lidar-guided cross-attention fusion for hyperspectral band selection and image classification,” IEEE Transactions on Geoscience and Remote Sensing, 2024

2024
[12]

Hyperspectral band selection: A review,

W. Sun and Q. Du, “Hyperspectral band selection: A review,” IEEE Geoscience and Remote Sensing Magazine , vol. 7, no. 2, pp. 118–139, 2019

2019
[13]

Attention is all you need,

A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” Advances in neural information processing systems , vol. 30, 2017

2017
[14]

Transformers in remote sensing: A survey,

A. A. Aleissaee, A. Kumar, R. M. Anwer, S. Khan, H. Cholakkal, G.-S. Xia, and F. S. Khan, “Transformers in remote sensing: A survey,” Remote Sensing, vol. 15, no. 7, p. 1860, 2023

2023
[15]

Spectralformer: Rethinking hyperspectral image classification with transformers,

D. Hong, Z. Han, J. Yao, L. Gao, B. Zhang, A. Plaza, and J. Chanussot, “Spectralformer: Rethinking hyperspectral image classification with transformers,” IEEE Transactions on Geoscience and Remote Sensing , vol. 60, pp. 1–15, 2021

2021
[16]

Hsi-bert: Hyperspectral image classification using the bidirectional encoder representation from transformers,

J. He, L. Zhao, H. Yang, M. Zhang, and W. Li, “Hsi-bert: Hyperspectral image classification using the bidirectional encoder representation from transformers,” IEEE Transactions on Geoscience and Remote Sensing , vol. 58, no. 1, pp. 165–178, 2019

2019
[17]

Mamba: Linear-time sequence modeling with selective state spaces,

A. Gu and T. Dao, “Mamba: Linear-time sequence modeling with selective state spaces,” arXiv preprint arXiv:2312.00752 , 2023

2023 arXiv
[18]

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,” arXiv preprint arXiv:2401.09417 , 2024

2024 arXiv
[19]

Learning compact and discriminative stacked autoencoder for hyperspectral image classification,

P. Zhou, J. Han, G. Cheng, and B. Zhang, “Learning compact and discriminative stacked autoencoder for hyperspectral image classification,” IEEE Transactions on Geoscience and Remote Sensing , vol. 57, no. 7, pp. 4823–4833, 2019. 17

2019
[20]

Convolutional neural networks for hyperspectral image classification,

S. Yu, S. Jia, and C. Xu, “Convolutional neural networks for hyperspectral image classification,” Neurocomputing, vol. 219, pp. 88–98, 2017

2017
[21]

Deep recurrent neural networks for hyperspectral image classification,

L. Mou, P. Ghamisi, and X. X. Zhu, “Deep recurrent neural networks for hyperspectral image classification,” IEEE Transactions on Geoscience and Remote Sensing , vol. 55, no. 7, pp. 3639–3655, 2017

2017
[22]

Generative adversarial networks for hyperspectral image classification,

L. Zhu, Y . Chen, P. Ghamisi, and J. A. Benediktsson, “Generative adversarial networks for hyperspectral image classification,” IEEE Transactions on Geoscience and Remote Sensing , vol. 56, no. 9, pp. 5046–5063, 2018

2018
[23]

Capsule networks for hyperspectral image classification,

M. E. Paoletti, J. M. Haut, R. Fernandez-Beltran, J. Plaza, A. Plaza, J. Li, and F. Pla, “Capsule networks for hyperspectral image classification,” IEEE Transactions on Geoscience and Remote Sensing , vol. 57, no. 4, pp. 2145–2160, 2018

2018
[24]

Hyperspectral image classification with deep learning models,

X. Yang, Y . Ye, X. Li, R. Y . Lau, X. Zhang, and X. Huang, “Hyperspectral image classification with deep learning models,” IEEE Transactions on Geoscience and Remote Sensing , vol. 56, no. 9, pp. 5408–5423, 2018

2018
[25]

Multi-scale 3d deep convolutional neural network for hyperspectral image classification,

M. He, B. Li, and H. Chen, “Multi-scale 3d deep convolutional neural network for hyperspectral image classification,” in 2017 IEEE International Conference on Image Processing (ICIP) . IEEE, 2017, pp. 3904–3908

2017
[26]

Deepvit: Towards deeper vision transformer,

D. Zhou, B. Kang, X. Jin, L. Yang, X. Lian, Z. Jiang, Q. Hou, and J. Feng, “Deepvit: Towards deeper vision transformer,” arXiv preprint arXiv:2103.11886, 2021

2021 arXiv
[27]

Tokens-to-token vit: Training vision transformers from scratch on imagenet,

L. Yuan, Y . Chen, T. Wang, W. Yu, Y . Shi, Z.-H. Jiang, F. E. Tay, J. Feng, and S. Yan, “Tokens-to-token vit: Training vision transformers from scratch on imagenet,” in Proceedings of the IEEE/CVF international conference on computer vision , 2021, pp. 558–567

2021
[28]

Levit: a vision transformer in convnet’s clothing for faster inference,

B. Graham, A. El-Nouby, H. Touvron, P. Stock, A. Joulin, H. J ´egou, and M. Douze, “Levit: a vision transformer in convnet’s clothing for faster inference,” in Proceedings of the IEEE/CVF international conference on computer vision , 2021, pp. 12 259–12 269

2021
[29]

Hyperspectral image transformer classification networks,

X. Yang, W. Cao, Y . Lu, and Y . Zhou, “Hyperspectral image transformer classification networks,” IEEE Transactions on Geoscience and Remote Sensing , vol. 60, pp. 1–15, 2022

2022
[30]

Multiscale convolutional transformer with center mask pretraining for hyperspectral image classification,

S. Jia and Y . Wang, “Multiscale convolutional transformer with center mask pretraining for hyperspectral image classification,” arXiv preprint arXiv:2203.04771, 2022

2022 arXiv
[31]

Spectral–spatial feature tokenization transformer for hyperspectral image classification,

L. Sun, G. Zhao, Y . Zheng, and Z. Wu, “Spectral–spatial feature tokenization transformer for hyperspectral image classification,” IEEE Transactions on Geoscience and Remote Sensing , vol. 60, pp. 1–14, 2022

2022
[32]

Hungry hungry hippos: Towards language modeling with state space models,

D. Y . Fu, T. Dao, K. K. Saab, A. W. Thomas, A. Rudra, and C. R ´e, “Hungry hungry hippos: Towards language modeling with state space models,” arXiv preprint arXiv:2212.14052 , 2022

2022 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Hyperspectral image super-resolution meets deep learning: A survey and perspective,

X. Wang, Q. Hu, Y . Cheng, and J. Ma, “Hyperspectral image super-resolution meets deep learning: A survey and perspective,” IEEE/CAA Journal of Automatica Sinica , vol. 10, no. 8, pp. 1668–1691, 2023

work page 2023

[2] [2]

Bands sensitive convolutional network for hyperspectral image classification,

L. Ran, Y . Zhang, W. Wei, and T. Yang, “Bands sensitive convolutional network for hyperspectral image classification,” in Proceedings of the International Conference on Internet Multimedia Computing and Service , 2016, pp. 268–272

work page 2016

[3] [3]

Ikeuchi, Computer vision: A reference guide

K. Ikeuchi, Computer vision: A reference guide . Springer, 2021

work page 2021

[4] [4]

The role of hyperspectral imaging: A literature review,

M. Mateen, J. Wen, M. A. Akbar et al., “The role of hyperspectral imaging: A literature review,” International Journal of Advanced Computer Science and Applications , vol. 9, no. 8, 2018

work page 2018

[5] [5]

A review on the combination of deep learning techniques with proximal hyperspectral images in agriculture,

J. G. A. Barbedo, “A review on the combination of deep learning techniques with proximal hyperspectral images in agriculture,” Computers and Electronics in Agriculture , vol. 210, p. 107920, 2023

work page 2023

[6] [6]

Automatic apple recognition based on the fusion of color and 3D feature for robotic fruit picking,

Y . Tao and J. Zhou, “Automatic apple recognition based on the fusion of color and 3D feature for robotic fruit picking,” Computers and Electronics in Agriculture, vol. 142, pp. 388–396, 2017

work page 2017

[7] [7]

K. P. Vadrevu, T. Le Toan, S. S. Ray, and C. O. Justice, Remote Sensing of Agriculture and Land Cover/Land Use Changes in South and Southeast Asian Countries . Springer, 2022

work page 2022

[8] [8]

Land use and land cover classification with hyperspectral data: A comprehensive review of methods, challenges and future directions,

M. A. Moharram and D. M. Sundaram, “Land use and land cover classification with hyperspectral data: A comprehensive review of methods, challenges and future directions,” Neurocomputing, 2023

work page 2023