Pith. sign in

REVIEW 3 major objections 5 minor 52 references

Hyperspectral Image Cross-Domain Object Detection Method based on Spectral-Spatial Feature Alignment

T0 review · 3 major / 5 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 The paper claims that hyperspectral object detectors can be transferred across sensors and scenes without target labels by aligning local spectral-spatial features, reporting 23.9% AP on a cross-domain ship-detection task.

desk verdict This is a plausible first step on a real task, but the near-zero baseline scores are not credible evidence until the baselines are adapted to HSI input fairly. read the letter →

arxiv 2411.16772 v1 pith:2YPVRUCV submitted 2024-11-25 cs.CV

Hongqi Zhang , He Sun , Hongmin Gao , Feng Han , Xu Sun , Lianru Gao , Bing Zhang This is my paper · ORCID
classification cs.CV
keywords hyperspectralobjectdetectioncross-domaindomainadaptationspectral-spatialfeaturesautoencoderspectralautocorrelationshipdataset
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 enough spectral detail to identify materials, but detectors trained on one sensor fail on another because spatial and spectral resolutions differ. This paper tries to close that gap without any labels in the target domain by assuming that the local spectral-spatial relationship between an object and its background stays the same across domains. It introduces a two-part network, SFA, that first extracts such local features with an autoencoder and then aligns their spectral autocorrelation matrices between source and target. On a ship-detection task moving from a coarse-resolution airborne dataset to a new fine-resolution port dataset, SFA reports 23.9% average precision while existing RGB-style cross-domain detectors score near zero. If the claim holds, it would make hyperspectral detectors transferable to new sensors and scenes without re-annotation.

What carries the argument

The load-bearing object is the pair of alignment modules. SSAM is an autoencoder (AE) trained with reconstruction loss plus an $\ell^1$ penalty on the third encoder layer, whose encoded feature map feeds a domain classifier via a gradient reversal layer to confuse source and target representations; this is meant to extract features encoding local shape and texture plus the spectral contrast between object and background. SACM then takes the same encoder features $F_S$ and $F_T$ from each domain and minimizes $\|F_T^T F_T - F_S^T F_S\|_F^2$, i.e., it aligns second-order spectral statistics so that objects of the same class exhibit similar spectral autocorrelation regardless of sensor. Together they replace the usual ImageNet backbone of a Faster-RCNN-style detector, whose RPN and ROI heads consume the aligned features.

What would settle it

Run the same M2SODAI→LWP experiment but replace SACM with a fixed random projection regularizer (or a simple feature-variance penalty) that does not use spectral structure; if AP stays near 23.9%, the spectral autocorrelation alignment itself is not the cause of the 7% gain. Alternatively, split the LWP test annotations by hull material and measure SACM's AP gain on homogeneous-material ships versus mixed-material ships; the method predicts a larger gain where materials are consistent.

Watch

Extended reading notes

Core claim

The paper's central claim is that unsupervised cross-domain object detection in hyperspectral images becomes tractable when the detector is constrained to align invariant local spectral-spatial features rather than global appearance. The authors argue that while global illumination, resolution, and spectral sampling vary between domains, the contrast between a man-made object and its immediate background persists in the spectral domain. They instantiate this in SFA, which couples a spectral-spatial alignment module (SSAM) — an autoencoder whose bottleneck features feed a domain classifier through a gradient reversal layer — with a spectral autocorrelation module (SACM) that penalizes differences between the source and target Gram matrices $F^T F$ of the extracted features. The reported result is 23.9% AP on M2SODAI→LWP, against ~0 AP for all compared UCOD baselines; ablations attribute 15.8% AP to SSAM and a further 7.0% to SACM. The paper also contributes the LWP dataset, 1,634 annotated hyperspectral images of a port at 0.1–0.2 m resolution.

Load-bearing premise

The premise is that objects of the same class are made of the same material, so their local spectral autocorrelations are transferable across domains — yet the paper's own target dataset contains ships of different materials (fishing ships of mixed materials, kelp ships of wood).

Editorial extensions

If this is right

  • If SFA generalizes, hyperspectral object detectors can be deployed on new sensors and flight altitudes without collecting target labels, a step toward operational remote sensing.
  • The reported ablation numbers imply that spectral autocorrelation alignment contributes a large, separable gain (7.0% AP) on top of spatial alignment, suggesting that spectral shift is a primary failure mode that future HSI adaptation methods should target explicitly.
  • Because all existing UCOD baselines collapse to zero AP on this task, the paper's result redefines the evaluation paradigm for cross-domain hyperspectral detection and makes the LWP dataset a benchmark for it.
  • The reverse direction (LWP→M2SODAI) yields only 1.71% AP, so the method's transferability is asymmetric and tied to the spatial scale and background complexity of the source domain.

Reading between the lines

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

  • The same-material premise underlying SACM is not tested by the paper's own data: the LWP description notes fishing ships of different materials and kelp ships of wood, so aligning spectral autocorrelations across such heterogeneous classes may be matching a statistical mixture rather than a physical material signature; the 7% gain might be an artifact of the MSE regularizer suppressing feature var
  • A direct test would be to hold out homogeneous-material classes (e.g., only metal-hulled ships) and compare SACM's gain on those against mixed-material classes; the method predicts a larger gain where materials match.
  • Since the paper attributes most of the AP to small and medium objects and notes that object size is entangled with spectral-spatial features, future work could try to decouple scale from spectral alignment, e.g., by aligning spectral statistics per object scale or per class.
  • The success of AE-based feature extraction suggests that self-supervised reconstruction losses may be more effective for HSI domain adaptation than the adversarial feature alignment used in RGB UCOD, a hypothesis that could be tested by plugging SSAM-style AEs into existing UCOD frameworks.
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 proposes SFA, an unsupervised cross-domain object detection method for hyperspectral images (HSI). It introduces a spectral-spatial alignment module (SSAM) based on an autoencoder to extract local spectral-spatial features, a domain classifier with gradient reversal for alignment, and a spectral autocorrelation module (SACM) that aligns spectral autocorrelation matrices across domains. The authors collect and annotate a new HSI dataset (Longwang Port, LWP) and evaluate on M2SODAI→LWP, reporting 23.9% AP against near-zero AP for existing UCOD baselines. Ablation studies attribute 15.8% AP to SSAM and 7.0% AP to SACM. The paper claims to be the first HCOD framework in the object detection community.

Significance. The paper tackles a genuinely new task—cross-domain object detection in hyperspectral imagery—and contributes a new annotated dataset, the LWP dataset. The proposed architecture explicitly models spectral and spatial domain shift, and the empirical result (23.9% AP on M2SODAI→LWP) is the first demonstration of a working HCOD system. The ablation study provides direct evidence for the contributions of the two modules. However, the significance is tempered by a major confound in the baseline comparison: the near-zero performance of existing UCOD methods may be an artifact of the band-expansion protocol that destroys their ImageNet-pretrained first-layer features. The reverse-direction experiment (LWP→M2SODAI) yields only 1.71% AP, which weakens the claim of a general HCOD solution. If the baseline issue is corrected and the SACM mechanism is validated against its stated premise, the paper would make a solid contribution.

major comments (3)
  1. [§4.1 and §3.2] The baseline comparison is confounded by the band-expansion procedure. The paper states that comparison methods keep all settings unchanged except for the number of input channels of the backbone, expanding source bands by copying or downsampling. For ImageNet-pretrained VGG16/ResNet-based detectors (DA-Faster, PT, MGADA, AT, MRT), replacing the first 3-channel convolutional layer with a 270-channel layer initialized randomly discards the pretrained low-level features on which these methods rely. The near-zero AP results in Table 1 are therefore consistent with a broken first layer rather than with a fundamental failure of UCOD methods to handle HSI domain shift. This weakens the headline claim that existing SOTA UCOD methods fail on HCOD. Please adapt baselines with an appropriate spectral input representation (e.g., PCA to 3 channels, selected-band composites, or fine-tuned spectral encoders) and report results under at least one matched protocol.
  2. [§3.4 and §3.5] The spectral autocorrelation module's premise conflicts with the dataset composition. Section 3.4 states the module is based on the hypothesis that 'if the same kind of object are made of same material, the spectral features should be invariant between domains.' However, Section 3.5 describes the LWP dataset as consisting of 'fishing ships made of different materials and kelp ships made of wood,' which directly violates the same-material assumption. Consequently, the 7.0% AP gain from SACM (Table 2) could result from an unintended regularizer on the learned feature autocorrelations rather than from physically meaningful spectral alignment. To support the mechanistic claim, please evaluate SACM against alternative feature-matching losses (e.g., MMD, CORAL, or simple feature norm matching) and analyze whether the aligned autocorrelation matrices correspond to spectrally similar materials.
  3. [Table 1, Table 2, and Supplementary Table 1] All reported metrics come from single runs with no error bars or repeated-seed statistics. On a small target test set (134 HSIs, 119 ships), detection AP is known to be noisy; the 23.9% AP result requires variance reporting (e.g., mean±std over at least 3 seeds) to assess its reliability. Additionally, the reverse cross-domain setting (LWP→M2SODAI) achieves only 1.71% AP in the supplementary material, which is effectively failure. The explanation based on spatial resolution differences is plausible but untested, and this asymmetry should be analyzed in the main text with failure cases rather than relegated to a supplementary note.
minor comments (5)
  1. [Eqs. (4) and (5)] The reconstruction loss expressions contain a stray '- +' symbol; the Frobenius norm term and the L1 term should be summed.
  2. [Algorithm 1] Algorithm 1 references equations incorrectly: line 5 should refer to Eq. (5) for the target feature extraction, and line 10 refers to Eq. (8) but the target domain classifier loss is given by Eq. (9). Please correct the line references.
  3. [§3.3] The sentence 'the loss involved Lr_s, Lr_s, Ld_s, and the Ld_t' repeats Lr_s twice; the first should likely be Lr_t, and the sentence should be rephrased for clarity.
  4. [Table 1] Column headers contain typos: 'APmidium', 'APmedium' should likely be 'APmedium', and 'ARmeduim' should be 'ARmedium'.
  5. [§1 and §5] The claim of being 'the first attempt in the object detection community' should be qualified with a more thorough discussion of prior hyperspectral domain adaptation works in other tasks, and the LWP dataset promised for release should include a public URL or availability statement.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: SFA's claims rest on held-out target AP and ablations, not on self-citation or fitted-parameter renaming.

full rationale

The paper's central claims are empirical and benchmark-grounded. The proposed SFA method is evaluated by AP on the held-out LWP validation set (Sec. 4.1, Table 1), and the contributions of SSAM and SACM are isolated by retraining with modules removed (Sec. 4.3, Table 2). The loss weights (ε=0.5, η=0.5, τ=0.2) and other hyperparameters are fixed, not fitted to the target test set. The SACM alignment loss in Eq. (10) is an optimization objective rather than a claimed prediction, and its 7.0% AP gain is an ablated measurement, not a quantity derived from the loss itself. No load-bearing step is justified by a self-citation; references to M2SODAI [19] and Detectron2 [48] are external resources, and no uniqueness theorem or prior-work ansatz is imported from the authors. The assumption that local spectral-spatial characteristics are invariant across domains is a stated premise, not a conclusion drawn from the premise, and the paper's dataset description (Sec. 3.5) even notes material differences that could challenge this premise. The baseline comparison protocol, which changes the backbone input from 3 to 270 channels for pretrained RGB detectors, may raise a fairness concern about whether the near-zero baseline AP is an artifact of broken first-layer features, but this is a benchmarking-design issue rather than circular reasoning: the paper does not define the target result in terms of the input or fit parameters so as to force the outcome. Overall, the derivation chain is self-contained with respect to the paper's own formulas, and the main evidence is externally measurable performance.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claim rests on domain-invariance assumptions that are stated but not proven, and on hand-set hyperparameters. The most serious issue is the same-material assumption, which conflicts with the dataset's own description.

free parameters (2)
  • Loss weights epsilon, eta, tau = 0.5, 0.5, 0.2
    Chosen by hand in Sec 3.2 to balance reconstruction, domain, and SACM losses; no tuning analysis.
  • GRL multiplier, beta, lambda = -0.5, 2.0, 0.25
    Default hyperparameters for gradient reversal and sigmoid focus loss in Sec 3.3; not justified by experiments.
assumptions (4)
  • domain assumption Local spectral-spatial characteristics remain invariant across different HSI domains
    Core premise stated in the abstract and Sec 3.3/3.4; illustrated by Figure 1 but not quantitatively verified.
  • domain assumption Objects of the same semantic class are made of the same material, so their spectral autocorrelations should match
    Explicit in Sec 3.4; contradicted by Sec 3.5 which says LWP contains fishing ships of different materials and kelp ships made of wood.
  • ad hoc to paper Band number matching by copying or downsampling source bands preserves spectral information needed for alignment
    Described at the end of Sec 3.2; no analysis of how interpolation or copying affects the spectral autocorrelation loss.
  • domain assumption Autoencoder reconstruction loss captures useful local spectral-spatial features
    Assumed in Sec 3.3 without comparison to alternative self-supervised pretext tasks.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hyperspectral Image Cross-Domain Object Detection Method based on Spectral-Spatial Feature Alignment." pith.science (2026). https://pith.science/paper/2YPVRUCV

@misc{pith2026241116772,
  author       = {Pith},
  title        = {Pith review of: Hyperspectral Image Cross-Domain Object Detection Method based on Spectral-Spatial Feature Alignment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2YPVRUCV}},
  note         = {Machine review of arXiv:2411.16772}
}
read the original abstract

With consecutive bands in a wide range of wavelengths, hyperspectral images (HSI) have provided a unique tool for object detection task. However, existing HSI object detection methods have not been fully utilized in real applications, which is mainly resulted by the difference of spatial and spectral resolution between the unlabeled target domain and a labeled source domain, i.e. the domain shift of HSI. In this work, we aim to explore the unsupervised cross-domain object detection of HSI. Our key observation is that the local spatial-spectral characteristics remain invariant across different domains. For solving the problem of domain-shift, we propose a HSI cross-domain object detection method based on spectral-spatial feature alignment, which is the first attempt in the object detection community to the best of our knowledge. Firstly, we develop a spectral-spatial alignment module to extract domain-invariant local spatial-spectral features. Secondly, the spectral autocorrelation module has been designed to solve the domain shift in the spectral domain specifically, which can effectively align HSIs with different spectral resolutions. Besides, we have collected and annotated an HSI dataset for the cross-domain object detection. Our experimental results have proved the effectiveness of HSI cross-domain object detection, which has firstly demonstrated a significant and promising step towards HSI cross-domain object detection in the object detection community.

Figures

Figures reproduced from arXiv: 2411.16772 by the authors.

Figure 1
Figure 1. Illustration of the invariant local spatial-spectral [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overview of SFA. the spectral and spatial domains. 2.3. HSI Object Detection Datasets The HOD task has been developed for decades, but there are no sufficient large-scale public datasets now. The most pop￾ular one is the Airport-Beach-Urban (ABU) dataset [22], which only includes 13 HSIs of a rather small spatial size of 100×100. Although they have been widely used for HOD, its detected object is too obvious and… view at source ↗
Figure 3
Figure 3. The overview of the SSAM. is mainly caused by the local spectral-spatial features being less compared with features of inter-domain changes. We argue the n-level layer of the encoder as the enn, and the operation of the AE is denoted by ae(.). Accordingly, the reconstructed loss L r s and the L r t can be written as follows: L r s = [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: The overview of the SACM. of FPN. When propagating forward, its features remain un￾changed. When performing the gradient back-propagation, the feature layer output by FPN is multiplied by a negative number (setting to -0.5 in the experiment) to obfuscate the local spec…
Figure 5
Figure 5. Figure 5: Illustration of the domain shift phenomenon in spectral-spatial dimension between M2SODAI and LWP datasets. The [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Qualitative results on the target domain. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: The feature map for the ablation experiment. [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: The bounding box distribution between M2SODAI [ [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 1
Figure 1. Figure 1: (h), the land background on the right side is mistak￾enly identified as part of the ship object. When using the M2SODAI as the source domain, we can leverage the cor￾responding annotation. Therefore, it enables the method to adapt to the sophisticated scene, and it can…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

52 extracted references · 52 canonical work pages

  1. [1]

    A broadband hyperspectral image sensor with high spatio-temporal reso- lution

    Liheng Bian, Zhen Wang, Yuzhe Zhang, Lianjie Li, Yinuo Zhang, Chen Yang, Wen Fang, Jiajun Zhao, Chunli Zhu, Qinghao Meng, Xuan Peng, and Jun Zhang. A broadband hyperspectral image sensor with high spatio-temporal reso- lution. Nature, 635(8037):73–81, 2024. 1

    work page 2024

  2. [2]

    Mask-guided spectral-wise transformer for efficient hyper- spectral image reconstruction

    Yuanhao Cai, Jing Lin, Xiaowan Hu, Haoqian Wang, Xin Yuan, Yulun Zhang, Radu Timofte, and Luc Van Gool. Mask-guided spectral-wise transformer for efficient hyper- spectral image reconstruction. In CVPR, pages 17502– 17511, 2022. 1

    work page 2022

  3. [3]

    Contrastive mean teacher for domain adaptive object detectors

    Shengcao Cao, Dhiraj Joshi, Liang-Yan Gui, and Yu-Xiong Wang. Contrastive mean teacher for domain adaptive object detectors. In CVPR, pages 23839–23848, 2023. 2

    work page 2023

  4. [4]

    Constrained band selection for hyperspectral imagery

    Chein-I Chang and Su Wang. Constrained band selection for hyperspectral imagery. TGRS, 44(6):1575–1585, 2006. 2

    work page 2006

  5. [5]

    Harmonizing transferability and discriminability for adapting object detectors

    Chaoqi Chen, Zebiao Zheng, Xinghao Ding, Yue Huang, and Qi Dou. Harmonizing transferability and discriminability for adapting object detectors. In CVPR, pages 8869–8878, 2020. 2

    work page 2020

  6. [6]

    Learning domain adaptive object detection with probabilistic teacher

    Meilin Chen, Weijie Chen, Shicai Yang, Jie Song, Xin- chao Wang, Lei Zhang, Yunfeng Yan, Donglian Qi, Yuet- ing Zhuang, Di Xie, et al. Learning domain adaptive object detection with probabilistic teacher. In ICML, pages 3040– 3055, 2022. 1, 2, 6, 7

    work page 2022

  7. [7]

    Domain adaptive faster r-cnn for object de- tection in the wild

    Yuhua Chen, Wen Li, Christos Sakaridis, Dengxin Dai, and Luc Van Gool. Domain adaptive faster r-cnn for object de- tection in the wild. In CVPR, pages 3339–3348, 2018. 1, 2, 3, 6, 7

    work page 2018

  8. [8]

    Un- biased mean teacher for cross-domain object detection

    Jinhong Deng, Wen Li, Yuhua Chen, and Lixin Duan. Un- biased mean teacher for cross-domain object detection. In CVPR, pages 4091–4101, 2021. 2

    work page 2021

Show all 52 references
  1. [9]

    Harmo- nious teacher for cross-domain object detection

    Jinhong Deng, Dongli Xu, Wen Li, and Lixin Duan. Harmo- nious teacher for cross-domain object detection. In CVPR, pages 23829–23838, 2023. 1, 2

    2023

  2. [11]

    D3t: Distinctive dual-domain teacher zigzagging across rgb- thermal gap for domain-adaptive object detection

    Dinh Phat Do, Taehoon Kim, Jaemin Na, Jiwon Kim, Keonho Lee, Kyunghwan Cho, and Wonjun Hwang. D3t: Distinctive dual-domain teacher zigzagging across rgb- thermal gap for domain-adaptive object detection. In CVPR, pages 23313–23322, 2024. 2

    2024

  3. [12]

    Asymmetric weighted logistic metric learn- ing for hyperspectral target detection

    Yanni Dong, Wenzhong Shi, Bo Du, Xiangyun Hu, and Liangpei Zhang. Asymmetric weighted logistic metric learn- ing for hyperspectral target detection. TCYB, 52(10):11093– 11106, 2022. 2

    2022

  4. [13]

    Unsupervised domain adaptation by backpropagation

    Yaroslav Ganin and Victor Lempitsky. Unsupervised domain adaptation by backpropagation. In ICML, pages 1180–1189,

  5. [14]

    Csda: Learning category-scale joint feature for domain adaptive object detection

    Changlong Gao, Chengxu Liu, Yujie Dun, and Xueming Qian. Csda: Learning category-scale joint feature for domain adaptive object detection. In ICCV, pages 11421–11430,

  6. [15]

    Cross domain object detection by target-perceived dual branch distillation

    Mengzhe He, Yali Wang, Jiaxi Wu, Yiru Wang, Hanqing Li, Bo Li, Weihao Gan, Wei Wu, and Yu Qiao. Cross domain object detection by target-perceived dual branch distillation. In CVPR, pages 9570–9580, 2022. 1

    2022

  7. [16]

    Surface plasmons-phonons for mid-infrared hyperspectral imaging

    Zhou Hong, Li Dongxiao, Ren Zhihao, Xu Cheng, Wang Lin-Fa, and Lee Chengkuo. Surface plasmons-phonons for mid-infrared hyperspectral imaging. Science Advances, 10 (22):eado3179, 2024. 1

    2024

  8. [17]

    Every pixel matters: Center-aware feature alignment for domain adaptive object detector

    Cheng-Chun Hsu, Yi-Hsuan Tsai, Yen-Yu Lin, and Ming- Hsuan Yang. Every pixel matters: Center-aware feature alignment for domain adaptive object detector. In ECCV, pages 733–748, 2020. 2

    2020

  9. [18]

    Progressive domain adaptation for object detection

    Han-Kai Hsu, Chun-Han Yao, Yi-Hsuan Tsai, Wei-Chih Hung, Hung-Yu Tseng, Maneesh Singh, and Ming-Hsuan Yang. Progressive domain adaptation for object detection. In WACV, pages 749–757, 2020. 2

    2020

  10. [19]

    M2sodai: Multi- modal maritime object detection dataset with rgb and hyper- spectral image sensors

    Jonggyu Jang, Sangwoo Oh, Youjin Kim, Dongmin Seo, Youngchol Choi, and Hyun Jong Yang. M2sodai: Multi- modal maritime object detection dataset with rgb and hyper- spectral image sensors. In NIPS, pages 53831–53843, 2023. 2, 3, 6, 7, 8

    2023

  11. [20]

    Decoupled adaptation for cross-domain object detec- tion

    Junguang Jiang, Baixu Chen, Jianmin Wang, and Mingsheng Long. Decoupled adaptation for cross-domain object detec- tion. In ICLR, 2021. 2

    2021

  12. [21]

    1 sparsity- regularized attention multiple-instance network for hyper- spectral target detection

    Changzhe Jiao, Chao Chen, Shuiping Gou, Xiuxiu Wang, Bo Yang, Xiaoying Chen, and Licheng Jiao. 1 sparsity- regularized attention multiple-instance network for hyper- spectral target detection. TCYB, 53(1):124–137, 2023. 2

    2023

  13. [22]

    Hyperspectral anomaly detec- tion with attribute and edge-preserving filters.TGRS, 55(10): 5600–5611, 2017

    Xudong Kang, Xiangping Zhang, Shutao Li, Kenli Li, Jun Li, and JonAtli Benediktsson. Hyperspectral anomaly detec- tion with attribute and edge-preserving filters.TGRS, 55(10): 5600–5611, 2017. 3

    2017

  14. [23]

    2pcnet: Two-phase consistency training for day-to-night unsupervised domain adaptive object detec- tion

    Mikhail Kennerley, Jian-Gang Wang, Bharadwaj Veeravalli, and Robby T Tan. 2pcnet: Two-phase consistency training for day-to-night unsupervised domain adaptive object detec- tion. In CVPR, pages 11484–11493, 2023. 2

    2023

  15. [24]

    Cat: Exploiting inter-class dynamics for domain adaptive object detection

    Mikhail Kennerley, Jian-Gang Wang, Bharadwaj Veeravalli, and Robby T Tan. Cat: Exploiting inter-class dynamics for domain adaptive object detection. In CVPR, pages 16541– 16550, 2024. 2

    2024

  16. [25]

    Cross-domain object detection for autonomous driving: A stepwise domain adaptative yolo approach

    Guofa Li, Zefeng Ji, Xingda Qu, Rui Zhou, and Dongpu Cao. Cross-domain object detection for autonomous driving: A stepwise domain adaptative yolo approach. TIV, 7(3):603– 615, 2022. 2

    2022

  17. [26]

    Quantization-aware deep optics for diffractive snapshot hyperspectral imaging

    Lingen Li, Lizhi Wang, Weitao Song, Lei Zhang, Zhiwei Xiong, and Hua Huang. Quantization-aware deep optics for diffractive snapshot hyperspectral imaging. In CVPR, pages 19780–19789, 2022. 1

    2022

  18. [27]

    Spectral enhanced rectangle transformer for hyperspectral image denoising

    Miaoyu Li, Ji Liu, Ying Fu, Yulun Zhang, and Dejing Dou. Spectral enhanced rectangle transformer for hyperspectral image denoising. In CVPR, pages 5805–5814, 2023. 1

    2023

  19. [28]

    Sigma: Semantic- complete graph matching for domain adaptive object detec- tion

    Wuyang Li, Xinyu Liu, and Yixuan Yuan. Sigma: Semantic- complete graph matching for domain adaptive object detec- tion. In CVPR, pages 5291–5300, 2022. 1

    2022

  20. [29]

    Target detection with unconstrained lin- ear mixture model and hierarchical denoising autoencoder in hyperspectral imagery

    Yunsong Li, Yanzi Shi, Keyan Wang, Bobo Xi, Jiaojiao Li, and Paolo Gamba. Target detection with unconstrained lin- ear mixture model and hierarchical denoising autoencoder in hyperspectral imagery. TIP, 31(1):1418–1432, 2022. 1

    2022

  21. [30]

    Cross-domain adaptive teacher for object detection

    Yu-Jhe Li, Xiaoliang Dai, Chih-Yao Ma, Yen-Cheng Liu, Kan Chen, Bichen Wu, Zijian He, Kris Kitani, and Peter Va- jda. Cross-domain adaptive teacher for object detection. In CVPR, pages 7581–7590, 2022. 1, 2, 6, 7

    2022

  22. [31]

    Decom- pose to adapt: Cross-domain object detection via feature dis- entanglement

    Dongnan Liu, Chaoyi Zhang, Yang Song, Heng Huang, Chenyu Wang, Michael Barnett, and Weidong Cai. Decom- pose to adapt: Cross-domain object detection via feature dis- entanglement. TMM, 25:1333–1344, 2022. 2

    2022

  23. [32]

    Confmix: Unsupervised domain adaptation for ob- ject detection via confidence-based mixing

    Giulio Mattolin, Luca Zanella, Elisa Ricci, and Yiming Wang. Confmix: Unsupervised domain adaptation for ob- ject detection via confidence-based mixing. In WACV, pages 423–433, 2023. 1

    2023

  24. [33]

    Active domain adaptation with false negative prediction for object detection

    Yuzuru Nakamura, Yasunori Ishii, and Takayoshi Yamashita. Active domain adaptation with false negative prediction for object detection. In CVPR, pages 28782–28792, 2024. 2

    2024

  25. [34]

    HTD- TS3: Weakly supervised hyperspectral target detection based on transformer via spectral–spatial similarity

    Haonan Qin, Weiying Xie, Yunsong Li, and Qian Du. HTD- TS3: Weakly supervised hyperspectral target detection based on transformer via spectral–spatial similarity. TNNLS, 35 (11):16816–16830, 2024. 2

    2024

  26. [35]

    Faster r-cnn: Towards real-time object detection with region proposal networks

    Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster r-cnn: Towards real-time object detection with region proposal networks. TPAMI, 39(6):1137–1149, 2016. 3

    2016

  27. [36]

    Prior-based domain adaptive object detec- tion for hazy and rainy conditions

    Vishwanath A Sindagi, Poojan Oza, Rajeev Yasarla, and Vishal M Patel. Prior-based domain adaptive object detec- tion for hazy and rainy conditions. InECCV, pages 763–780,

  28. [37]

    Adapting object detec- tors with conditional domain normalization

    Peng Su, Kun Wang, Xingyu Zeng, Shixiang Tang, Dapeng Chen, Di Qiu, and Xiaogang Wang. Adapting object detec- tors with conditional domain normalization. In ECCV, pages 403–419, 2020. 2

    2020

  29. [38]

    Mega-cda: Memory guided attention for category-aware unsupervised domain adaptive object detection

    Vibashan Vs, Vikram Gupta, Poojan Oza, Vishwanath A Sindagi, and Vishal M Patel. Mega-cda: Memory guided attention for category-aware unsupervised domain adaptive object detection. In CVPR, pages 4516–4526, 2021. 1

    2021

  30. [39]

    Meta-uda: Unsupervised domain adap- tive thermal object detection using meta-learning

    Vibashan Vs, Domenick Poster, Suya You, Shuowen Hu, and Vishal M Patel. Meta-uda: Unsupervised domain adap- tive thermal object detection using meta-learning. In WACV, pages 1412–1423, 2022. 2

    2022

  31. [40]

    Towards on- line domain adaptive object detection

    Vibashan VS, Poojan Oza, and Vishal M Patel. Towards on- line domain adaptive object detection. In WACV, pages 478– 488, 2023. 1

    2023

  32. [41]

    Generalized autoencoder: A neural network framework for dimensionality reduction

    Wei Wang, Yan Huang, Yizhou Wang, and Liang Wang. Generalized autoencoder: A neural network framework for dimensionality reduction. In CVPR, pages 490–497, 2014. 2

    2014

  33. [42]

    In2set: Intra-inter similarity exploit- ing transformer for dual-camera compressive hyperspectral imaging

    Xin Wang, Lizhi Wang, Xiangtian Ma, Maoqing Zhang, Lin Zhu, and Hua Huang. In2set: Intra-inter similarity exploit- ing transformer for dual-camera compressive hyperspectral imaging. In CVPR, pages 24881–24891, 2024. 1

    2024

  34. [43]

    Instance-invariant domain adaptive object detection via pro- gressive disentanglement

    Aming Wu, Yahong Han, Linchao Zhu, and Yi Yang. Instance-invariant domain adaptive object detection via pro- gressive disentanglement. TPAMI, 44(8):4178–4193, 2021. 2

    2021

  35. [44]

    Exploring categorical regularization for domain adap- tive object detection

    Chang-Dong Xu, Xing-Ran Zhao, Xin Jin, and Xiu-Shen Wei. Exploring categorical regularization for domain adap- tive object detection. In CVPR, pages 11724–11733, 2020. 2

    2020

  36. [45]

    Cross-domain detection via graph-induced prototype alignment

    Minghao Xu, Hang Wang, Bingbing Ni, Qi Tian, and Wenjun Zhang. Cross-domain detection via graph-induced prototype alignment. In CVPR, pages 12355–12364, 2020. 2

    2020

  37. [46]

    Multi-view adversarial discriminator: Mine the non-causal factors for object detection in unseen domains

    Mingjun Xu, Lingyun Qin, Weijie Chen, Shiliang Pu, and Lei Zhang. Multi-view adversarial discriminator: Mine the non-causal factors for object detection in unseen domains. In CVPR, pages 8103–8112, 2023. 2

    2023

  38. [47]

    Sc-uda: Style and content gaps aware unsupervised domain adaptation for object de- tection

    Fuxun Yu, Di Wang, Yinpeng Chen, Nikolaos Karianakis, Tong Shen, Pei Yu, Dimitrios Lymberopoulos, Sidi Lu, Weisong Shi, and Xiang Chen. Sc-uda: Style and content gaps aware unsupervised domain adaptation for object de- tection. In WACV, pages 382–391, 2022. 1

    2022

  39. [48]

    Detectron2

    Wu Yuxin, Kirillov Alexander, Massa Francisco, Lo Wan- Yen, and Girshick Ross. Detectron2. https://github. com/facebookresearch/detectron2, 2019. 6

    2019

  40. [49]

    Discrimina- tive multiple instance hyperspectral target characterization

    Alina Zare, Changzhe Jiao, and Taylor Glenn. Discrimina- tive multiple instance hyperspectral target characterization. TPAMI, 40(10):2342–2354, 2017. 2

    2017

  41. [50]

    Quantum- inspired spectral-spatial pyramid network for hyperspectral image classification

    Jie Zhang, Yongshan Zhang, and Yicong Zhou. Quantum- inspired spectral-spatial pyramid network for hyperspectral image classification. In CVPR, pages 9925–9934, 2023. 1

    2023

  42. [51]

    Task-specific inconsistency alignment for domain adaptive object detection

    Liang Zhao and Limin Wang. Task-specific inconsistency alignment for domain adaptive object detection. In CVPR, pages 14217–14226, 2022. 2

    2022

  43. [52]

    Masked retraining teacher- student framework for domain adaptive object detection

    Zijing Zhao, Sitong Wei, Qingchao Chen, Dehui Li, Yifan Yang, Yuxin Peng, and Yang Liu. Masked retraining teacher- student framework for domain adaptive object detection. In ICCV, pages 19039–19049, 2023. 1, 2, 6, 7

    2023

  44. [53]

    Multi-granularity alignment domain adaptation for object detection

    Wenzhang Zhou, Dawei Du, Libo Zhang, Tiejian Luo, and Yanjun Wu. Multi-granularity alignment domain adaptation for object detection. In CVPR, pages 9581–9590, 2022. 1, 2, 6, 7 Hyperspectral Image Cross-Domain Object Detection Method based on Spectral-Spatial Feature Alignment ...

    2022

Pith tools

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