pith. sign in

arxiv: 2606.09167 · v1 · pith:LMMHJUDAnew · submitted 2026-06-08 · 💻 cs.CV

Vision-Language Guided Hyperspectral Object Tracking via Semantics Fusion and Contextual Template Updating

Rui Yao , Yuhong Zhang , Kunyang Sun , Hancheng Zhu , Jiaqi Zhao , Zhiwen Shao , Abdulmotaleb El Saddik This is my paper

Pith reviewed 2026-06-27 16:58 UTC · model grok-4.3

classification 💻 cs.CV
keywords hyperspectral object trackingvision-language fusionband selectiontemplate updatingstate space modelmultimodal tracking
0
0 comments X

The pith

VLHTrack uses LLM descriptions to select useful spectral bands and dynamically update templates for hyperspectral tracking.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper claims that hyperspectral object tracking suffers from too many redundant spectral bands and from target appearance changes over time. It introduces a framework that first maps language model descriptions of the target to a smaller set of discriminative bands, then fuses the resulting visual features with language embeddings, and finally uses a state-space model to evolve the template representation frame by frame. If correct, the approach would let trackers exploit the full spectral richness of hyperspectral video without being overwhelmed by band count or by deformation. Experiments on the HOT2023 and HOT2024 benchmarks are presented as evidence that the combined modules exceed prior methods.

Core claim

VLHTrack establishes a semantic-to-spectral mapping in the Language-Guided Band Selection Module that uses LLM-generated object descriptions to accentuate discriminative bands, integrates visual and linguistic embeddings through a Multi-Modal Vision-Language Fusion Module, and applies selective state-space modeling in the Dynamic Template Update with Mamba module to evolve template features according to temporal context, yielding higher tracking accuracy than existing approaches on HOT2023 and HOT2024.

What carries the argument

Language-Guided Band Selection Module that creates a semantic-to-spectral mapping from LLM descriptions to reduce spectral redundancy.

If this is right

  • Spectral redundancy is reduced so that models generalize better across hyperspectral videos.
  • Cross-modal representations become coherent enough to improve robustness under occlusion and illumination change.
  • Template features evolve efficiently across long sequences without explicit deformation modeling.
  • Performance gains appear on both HOT2023 and HOT2024 benchmarks relative to prior trackers.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same language-to-band mapping idea could be tested on other spectral imaging tasks such as material classification.
  • Replacing the Mamba component with a different temporal model would isolate how much the state-space update itself contributes.
  • If the LLM descriptions contain systematic biases for certain object classes, tracking accuracy on those classes would degrade first.

Load-bearing premise

Large language model descriptions of objects can be mapped reliably onto the spectral bands that best distinguish the target without selection errors or dataset bias.

What would settle it

A controlled test in which the band-selection module is replaced by random or full-band input and tracking accuracy on HOT2023 or HOT2024 either stays the same or drops.

Figures

Figures reproduced from arXiv: 2606.09167 by Abdulmotaleb El Saddik, Hancheng Zhu, Jiaqi Zhao, Kunyang Sun, Rui Yao, Yuhong Zhang, Zhiwen Shao.

Figure 1
Figure 1. Figure 1: Representative HSV sequences with LLM-generated [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The overall architecture of VLHTrack. VLHTrack consists of LBSM, multi-modal vision–language fusion, Dynamic [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Detailed design of the proposed LBSM. encoder layer are then used as the input to the tracking head for subsequent prediction. This bidirectional flow enables semantic-aware fusion, ensuring that the learned representa￾tions remain highly target-aware. B. Language-Guided Band Selection Module (LBSM) Hyperspectral images capture fine-grained spectral signa￾tures that are intrinsically linked to a target’s p… view at source ↗
Figure 4
Figure 4. Figure 4: Detailed design of the proposed DTUM. where FFNl(·) denotes the feed-forward sublayer in the l-th transformer encoder block, and Eel t represents its output. For subsequent layers, the encoder takes Eel−1 t as input, enabling hierarchical refinement of cross-modal representations across depth. This design effectively captures both spectral and semantic cues, ensuring stable and effective synergy between vi… view at source ↗
Figure 5
Figure 5. Figure 5: Precision and success plots comparison with SOTA [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Precision and success plots comparison with SOTA [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: The precision and success of VLHTrack and other [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Our proposed VLHTrack is qualitatively compared with PHTrack, SENSE, SPIRIT, Trans-DAT, SSTtrack, EVPTrack, [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
read the original abstract

Hyperspectral object tracking (HOT) leverages the rich spectral information provided by hyperspectral videos (HSVs), offering substantial potential for object tracking. However, efficiently extracting and exploiting spectral information from redundant spectral bands remains a fundamental challenge, which severely limits model generalization and tracking performance. Moreover, in dynamic scenes, targets often experience drastic appearance variations due to factors such as occlusion and illumination changes. These variations lead to large deformations between the current frame and the template. Such discrepancies pose major challenges for existing temporal modeling approaches. In this work, we propose VLHTrack, a novel hyperspectral vision-language (VL) joint tracking framework. Specifically, we incorporate language priors to address the fundamental challenge of spectral redundancy by designing a Language-Guided Band Selection Module (LBSM). By leveraging Large Language Model (LLM) descriptions, LBSM establishes a semantic-to-spectral mapping that mitigates redundancy and accentuates discriminative spectral features. A Multi-Modal Vision-Language Fusion Module is then employed to seamlessly integrate visual and linguistic embeddings, harnessing their complementary advantages to learn coherent cross-modal representations. To address target deformation in long-term sequences, we propose a dynamic update template feature strategy implemented via the Dynamic Template Update with Mamba (DTUM) module. By leveraging selective state space modeling, DTUM learns inter-frame dependencies to update template feature, ensuring efficient template feature evolution guided by temporal context. Experiments on HOT2023 and HOT2024 demonstrate that VLHTrack outperforms state-of-the-art (SOTA) methods.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The paper proposes VLHTrack, a novel hyperspectral vision-language joint tracking framework. It introduces the Language-Guided Band Selection Module (LBSM) that leverages LLM descriptions to establish semantic-to-spectral mappings for mitigating spectral redundancy, a Multi-Modal Vision-Language Fusion Module for integrating visual and linguistic embeddings, and the Dynamic Template Update with Mamba (DTUM) module for updating template features using selective state space modeling to handle target deformations. The central claim is that VLHTrack outperforms state-of-the-art methods on the HOT2023 and HOT2024 datasets.

Significance. If the empirical results hold, this work would be significant for advancing hyperspectral object tracking by demonstrating how language priors from LLMs can guide band selection to reduce redundancy and how Mamba-based modeling can address long-term template updates. The approach combines timely elements from vision-language models and state space models in a domain where spectral information is underutilized.

major comments (1)
  1. [Abstract] Abstract: The assertion that 'Experiments on HOT2023 and HOT2024 demonstrate that VLHTrack outperforms state-of-the-art (SOTA) methods' is not supported by any quantitative scores, success/precision plots, ablation tables, or implementation details. This absence makes it impossible to verify the central claim or assess the contribution of the LBSM, fusion module, or DTUM.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback on the abstract. We address the major comment below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The assertion that 'Experiments on HOT2023 and HOT2024 demonstrate that VLHTrack outperforms state-of-the-art (SOTA) methods' is not supported by any quantitative scores, success/precision plots, ablation tables, or implementation details. This absence makes it impossible to verify the central claim or assess the contribution of the LBSM, fusion module, or DTUM.

    Authors: We agree that the abstract would be strengthened by including key quantitative results to directly support the performance claim. The full manuscript contains the requested elements (success/precision plots, ablation tables, and implementation details) in the Experiments section. To address this point, we will revise the abstract to incorporate representative metrics, such as success rate and precision scores on HOT2023 and HOT2024 along with the margin of improvement over prior SOTA methods. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical architecture evaluated on public benchmarks

full rationale

The paper proposes VLHTrack as an empirical framework consisting of LBSM (using LLM descriptions for band selection), multi-modal fusion, and DTUM (Mamba-based template update). No equations, derivations, or predictions are presented that reduce to fitted parameters or self-definitions by construction. Central claims rest on experimental outperformance on HOT2023/HOT2024, which are external public benchmarks. No self-citation chains or uniqueness theorems are invoked as load-bearing in the provided text. This is a standard architecture proposal with independent empirical content.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard deep-learning assumptions plus two domain-specific premises about language-to-spectrum mapping and selective state-space modeling; no new physical entities are postulated.

free parameters (1)
  • neural network hyperparameters and learned weights
    All modules are neural networks whose parameters are fitted to training data; exact count and values not stated in abstract.
axioms (2)
  • domain assumption LLM descriptions provide accurate and unbiased semantic information usable for spectral band selection
    Invoked by the Language-Guided Band Selection Module description.
  • domain assumption Mamba selective state-space modeling can capture useful inter-frame dependencies for template evolution
    Invoked by the Dynamic Template Update with Mamba module description.

pith-pipeline@v0.9.1-grok · 5827 in / 1375 out tokens · 20105 ms · 2026-06-27T16:58:24.588510+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

73 extracted references · 4 linked inside Pith

  1. [1]

    Siamese local and global networks for robust face tracking,

    Y . Qi, S. Zhang, F. Jiang, H. Zhou, D. Tao, and X. Li, “Siamese local and global networks for robust face tracking,”IEEE Transactions on Image Processing, vol. 29, pp. 9152–9164, 2020

    2020

  2. [2]

    Multi- modality sensing and data fusion for multi-vehicle detection,

    D. Roy, Y . Li, T. Jian, P. Tian, K. Chowdhury, and S. Ioannidis, “Multi- modality sensing and data fusion for multi-vehicle detection,”IEEE Transactions on Multimedia, vol. 25, pp. 2280–2295, 2023

    2023

  3. [3]

    Adversarial geometric attacks for 3d point cloud object tracking,

    R. Yao, A. Zhang, Y . Zhou, J. Zhao, B. Liu, and A. El Saddik, “Adversarial geometric attacks for 3d point cloud object tracking,”IEEE Transactions on Multimedia, 2025

    2025

  4. [4]

    Deep learning for 3d point clouds: A survey,

    Y . Guo, H. Wang, Q. Hu, H. Liu, L. Liu, and M. Bennamoun, “Deep learning for 3d point clouds: A survey,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 12, pp. 4338–4364, 2021

    2021

  5. [5]

    Visual tracking: An experimental survey,

    A. W. M. Smeulders, D. M. Chu, R. Cucchiara, S. Calderara, A. De- hghan, and M. Shah, “Visual tracking: An experimental survey,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, no. 7, pp. 1442–1468, 2014

    2014

  6. [6]

    Deep feature-based hyperspectral object tracking: An experimental survey and outlook,

    Y . Wang, X. Li, X. Yang, F. Ge, B. Wei, L. Li, and S. Yue, “Deep feature-based hyperspectral object tracking: An experimental survey and outlook,”Remote Sensing, vol. 17, no. 4, p. 645, 2025

    2025

  7. [7]

    Ssf-net: Spatial-spectral fusion network with spectral angle awareness for hyperspectral object tracking,

    H. Wang, W. Li, X.-G. Xia, Q. Du, and J. Tian, “Ssf-net: Spatial-spectral fusion network with spectral angle awareness for hyperspectral object tracking,”IEEE Transactions on Image Processing, 2025

    2025

  8. [8]

    Tracking via object reflectance using a hyperspectral video camera,

    H. V . Nguyen, A. Banerjee, and R. Chellappa, “Tracking via object reflectance using a hyperspectral video camera,” in2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops, 2010, pp. 44–51

    2010

  9. [9]

    Machine learning based hy- perspectral image analysis: a survey,

    U. B. Gewali, S. T. Monteiro, and E. Saber, “Machine learning based hy- perspectral image analysis: a survey,”arXiv preprint arXiv:1802.08701, 2018

    Pith/arXiv arXiv 2018

  10. [10]

    High-speed tracking with kernelized correlation filters,

    J. F. Henriques, R. Caseiro, P. Martins, and J. Batista, “High-speed tracking with kernelized correlation filters,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 37, no. 3, pp. 583–596, 2015

    2015

  11. [11]

    Spatial–spectral weighted and regularized tensor sparse correlation filter for object tracking in hyper- spectral videos,

    Z. Hou, W. Li, J. Zhou, and R. Tao, “Spatial–spectral weighted and regularized tensor sparse correlation filter for object tracking in hyper- spectral videos,”IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1–12, 2022

    2022

  12. [12]

    Material based object tracking in hyperspectral videos,

    F. Xiong, J. Zhou, and Y . Qian, “Material based object tracking in hyperspectral videos,”IEEE Transactions on Image Processing, vol. 29, pp. 3719–3733, 2020

    2020

  13. [13]

    Hyper- spectral object tracking with dual-stream prompt,

    R. Yao, L. Zhang, Y . Zhou, H. Zhu, J. Zhao, and Z. Shao, “Hyper- spectral object tracking with dual-stream prompt,”IEEE Transactions on Geoscience and Remote Sensing, vol. 63, pp. 1–12, 2025

    2025

  14. [14]

    Historical object-aware prompt learning for universal hyperspectral object tracking,

    L. Zhang, R. Yao, Y . Zhang, Y . Zhou, F. Hu, J. Zhao, and Z. Shao, “Historical object-aware prompt learning for universal hyperspectral object tracking,”ACM Transactions on Multimedia Computing, Com- munications and Applications, 2025

    2025

  15. [15]

    Ssttrack: A unified hyperspectral video tracking framework via modeling spectral-spatial-temporal conditions,

    Y . Chen, Q. Yuan, Y . Tang, Y . Xiao, J. He, T. Han, Z. Liu, and L. Zhang, “Ssttrack: A unified hyperspectral video tracking framework via modeling spectral-spatial-temporal conditions,”Information Fusion, vol. 114, p. 102658, 2025

    2025

  16. [16]

    Learning a deep ensemble network with band importance for hyperspectral object tracking,

    Z. Li, F. Xiong, J. Zhou, J. Lu, and Y . Qian, “Learning a deep ensemble network with band importance for hyperspectral object tracking,”IEEE Transactions on Image Processing, vol. 32, pp. 2901–2914, 2023

    2023

  17. [17]

    Material- guided multiview fusion network for hyperspectral object tracking,

    Z. Li, F. Xiong, J. Zhou, J. Lu, Z. Zhao, and Y . Qian, “Material- guided multiview fusion network for hyperspectral object tracking,” IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1–15, 2024

    2024

  18. [18]

    Siambag: Band attention grouping- based siamese object tracking network for hyperspectral videos,

    W. Li, Z. Hou, J. Zhou, and R. Tao, “Siambag: Band attention grouping- based siamese object tracking network for hyperspectral videos,”IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1–12, 2023. 12 XXXX IEEE TRANSACTIONS ON MULTIMEDIA, VOL. XX, XXXX

    2023

  19. [19]

    Language-guided dual-modal local correspondence for single object tracking,

    J. Yu, Z. Cai, Y . Li, L. Wang, F. Gao, and Y . Yu, “Language-guided dual-modal local correspondence for single object tracking,”IEEE Transactions on Multimedia, vol. 26, pp. 10 637–10 650, 2024

    2024

  20. [20]

    One-stream vision-language memory network for object tracking,

    H. Zhang, J. Wang, J. Zhang, T. Zhang, and B. Zhong, “One-stream vision-language memory network for object tracking,”IEEE Transac- tions on Multimedia, vol. 26, pp. 1720–1730, 2024

    2024

  21. [21]

    Divert more attention to vision- language tracking,

    M. Guo, Z. Zhang, H. Fan, and L. Jing, “Divert more attention to vision- language tracking,”Advances in Neural Information Processing Systems, vol. 35, pp. 4446–4460, 2022

    2022

  22. [22]

    Siamese natural language tracker: Tracking by natural language descriptions with siamese track- ers,

    Q. Feng, V . Ablavsky, Q. Bai, and S. Sclaroff, “Siamese natural language tracker: Tracking by natural language descriptions with siamese track- ers,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 5851–5860

    2021

  23. [23]

    Hyperspectral object tracking with context-aware learning and category consistency,

    Y . Wang, S. Mei, M. Ma, Y . Liu, T. Gao, and H. Han, “Hyperspectral object tracking with context-aware learning and category consistency,” IEEE Transactions on Geoscience and Remote Sensing, 2025

    2025

  24. [24]

    Gpt-4 technical report,

    J. Achiam, S. Adler, S. Agarwal, L. Ahmad, I. Akkaya, F. L. Aleman, D. Almeida, J. Altenschmidt, S. Altman, S. Anadkatet al., “Gpt-4 technical report,”arXiv preprint arXiv:2303.08774, 2023

    Pith/arXiv arXiv 2023

  25. [25]

    Hyperspectral mamba for hyperspectral object tracking,

    L. Gao, Y . Zhang, Y . Jiang, W. Xie, and Y . Li, “Hyperspectral mamba for hyperspectral object tracking,” 2025. [Online]. Available: https://arxiv.org/abs/2509.08265

    arXiv 2025

  26. [26]

    Improving visual object tracking through visual prompting,

    S.-F. Chen, J.-C. Chen, I.-H. Jhuo, and Y .-Y . Lin, “Improving visual object tracking through visual prompting,”IEEE Transactions on Mul- timedia, 2025

    2025

  27. [27]

    Spirit: Spectral awareness interaction network with dynamic template for hyperspectral object tracking,

    Y . Chen, Q. Yuan, Y . Tang, Y . Xiao, J. He, and L. Zhang, “Spirit: Spectral awareness interaction network with dynamic template for hyperspectral object tracking,”IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1–16, 2024

    2024

  28. [28]

    Object tracking in hyperspectral videos with convolutional features and kernelized correla- tion filter,

    K. Qian, J. Zhou, F. Xiong, H. Zhou, and J. Du, “Object tracking in hyperspectral videos with convolutional features and kernelized correla- tion filter,” inInternational conference on smart multimedia. Springer, 2018, pp. 308–319

    2018

  29. [29]

    Siamhyper: Learning a hyperspectral object tracker from an rgb-based tracker,

    Z. Liu, X. Wang, Y . Zhong, M. Shu, and C. Sun, “Siamhyper: Learning a hyperspectral object tracker from an rgb-based tracker,”IEEE Trans- actions on Image Processing, vol. 31, pp. 7116–7129, 2022

    2022

  30. [30]

    Bae-net: A band attention aware ensemble network for hyperspectral object tracking,

    Z. Li, F. Xiong, J. Zhou, J. Wang, J. Lu, and Y . Qian, “Bae-net: A band attention aware ensemble network for hyperspectral object tracking,” in 2020 IEEE International Conference on Image Processing (ICIP), 2020, pp. 2106–2110

    2020

  31. [31]

    Hy-tracker: A novel framework for enhancing efficiency and accuracy of object tracking in hyperspectral videos,

    M. A. Islam, W. Xing, J. Zhou, Y . Gao, and K. K. Paliwal, “Hy-tracker: A novel framework for enhancing efficiency and accuracy of object tracking in hyperspectral videos,”IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1–14, 2024

    2024

  32. [32]

    Tmtnet: A transformer-based multimodality information transfer network for hyperspectral object tracking,

    C. Zhao, H. Liu, N. Su, C. Xu, Y . Yan, and S. Feng, “Tmtnet: A transformer-based multimodality information transfer network for hyperspectral object tracking,”Remote Sensing, vol. 15, no. 4, p. 1107, 2023

    2023

  33. [33]

    Tftn: A transformer-based fusion tracking framework of hyperspectral and rgb,

    C. Zhao, H. Liu, N. Su, and Y . Yan, “Tftn: A transformer-based fusion tracking framework of hyperspectral and rgb,”IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1–15, 2022

    2022

  34. [34]

    Htacpe: A hybrid transformer with adaptive content and position embedding for sample learning efficiency of hyperspectral tracker,

    Y . Wang, S. Mei, M. Ma, Y . Liu, and Y . Su, “Htacpe: A hybrid transformer with adaptive content and position embedding for sample learning efficiency of hyperspectral tracker,”IEEE Transactions on Multimedia, vol. 27, pp. 2384–2398, 2025

    2025

  35. [35]

    Profit: A prompt-guided frequency-aware filtering and template- enhanced interaction framework for hyperspectral video tracking,

    Y . Chen, Q. Yuan, Y . Tang, X. Wang, Y . Xiao, J. He, Z. Lihe, and X. Jin, “Profit: A prompt-guided frequency-aware filtering and template- enhanced interaction framework for hyperspectral video tracking,”IS- PRS Journal of Photogrammetry and Remote Sensing, vol. 226, pp. 164– 186, 2025

    2025

  36. [36]

    Siamohot: A lightweight dual siamese network for onboard hyperspectral object tracking via joint spatial-spectral knowledge distillation,

    C. Sun, X. Wang, Z. Liu, Y . Wan, L. Zhang, and Y . Zhong, “Siamohot: A lightweight dual siamese network for onboard hyperspectral object tracking via joint spatial-spectral knowledge distillation,”IEEE Trans- actions on Geoscience and Remote Sensing, vol. 61, pp. 1–12, 2023

    2023

  37. [37]

    Hotmoe: Exploring sparse mixture-of-experts for hyperspectral object tracking,

    W. Sun, Y . Tan, J. Li, S. Hou, X. Li, Y . Shao, Z. Wang, and B. Song, “Hotmoe: Exploring sparse mixture-of-experts for hyperspectral object tracking,”IEEE Transactions on Multimedia, vol. 27, pp. 4072–4083, 2025

    2025

  38. [38]

    Hyperspectral object tracking via band and context refinement network,

    J. Zhang, Z. Zheng, K. Ni, N. Huang, Q. Liu, and P. Liu, “Hyperspectral object tracking via band and context refinement network,”Remote Sensing, vol. 17, no. 22, 2025

    2025

  39. [39]

    Supervised embedded methods for hyperspectral band selection,

    Y . Zimmer, O. Lindenbaum, and O. Glickman, “Supervised embedded methods for hyperspectral band selection,” 2025. [Online]. Available: https://arxiv.org/abs/2401.11420

    arXiv 2025

  40. [40]

    Enhancing vision-language tracking by effectively con- verting textual cues into visual cues,

    X. Feng, D. Zhang, S. Hu, X. Li, M. Wu, J. Zhang, X. Chen, and K. Huang, “Enhancing vision-language tracking by effectively con- verting textual cues into visual cues,” inICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2025, pp. 1–5

    2025

  41. [41]

    Vision-language tracking with clip and interactive prompt learning,

    H. Zhu, Q. Lu, L. Xue, P. Zhang, and G. Yuan, “Vision-language tracking with clip and interactive prompt learning,”IEEE Transactions on Intelligent Transportation Systems, vol. 26, no. 3, pp. 3659–3670, 2025

    2025

  42. [42]

    Learning transferable visual models from natural language supervision,

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

    2021

  43. [43]

    Memvlt: Vision-language tracking with adaptive memory-based prompts,

    X. Feng, X. Li, S. Hu, D. Zhang, J. Zhang, X. Chen, K. Huang et al., “Memvlt: Vision-language tracking with adaptive memory-based prompts,”Advances in Neural Information Processing Systems, vol. 37, pp. 14 903–14 933, 2024

    2024

  44. [44]

    Dtvlt: A multi-modal diverse text benchmark for visual language tracking based on llm,

    X. Li, S. Hu, X. Feng, D. Zhang, M. Wu, J. Zhang, and K. Huang, “Dtvlt: A multi-modal diverse text benchmark for visual language tracking based on llm,”arXiv preprint arXiv:2410.02492, 2024

    arXiv 2024

  45. [45]

    How texts help? a fine- grained evaluation to reveal the role of language in vision-language tracking,

    Li, Xuchen and Hu, Shiyu and Feng, Xiaokun and Zhang, Dailing and Wu, Meiqi and Zhang, Jing and Huang, Kaiqi, “How texts help? a fine- grained evaluation to reveal the role of language in vision-language tracking,”arXiv preprint arXiv:2411.15600, 2024

    arXiv 2024

  46. [46]

    Citetracker: Correlating image and text for visual tracking,

    X. Li, Y . Huang, Z. He, Y . Wang, H. Lu, and M.-H. Yang, “Citetracker: Correlating image and text for visual tracking,” in2023 IEEE/CVF International Conference on Computer Vision (ICCV), 2023, pp. 9940– 9949

    2023

  47. [47]

    Mamba adapter: Efficient multi-modal fusion for vision-language tracking,

    L. Shi, B. Zhong, Q. Liang, X. Hu, Z. Mo, and S. Song, “Mamba adapter: Efficient multi-modal fusion for vision-language tracking,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 35, no. 9, pp. 9300–9311, 2025

    2025

  48. [48]

    Language-aware domain generalization network for cross-scene hyperspectral image classification,

    Y . Zhang, M. Zhang, W. Li, S. Wang, and R. Tao, “Language-aware domain generalization network for cross-scene hyperspectral image classification,”IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1–12, 2023

    2023

  49. [49]

    Text-driven adaptive semantic alignment network for cross-scene hyperspectral image classification,

    W. Wang, F. Liu, H. Zhu, and L. Xiao, “Text-driven adaptive semantic alignment network for cross-scene hyperspectral image classification,” IEEE Transactions on Geoscience and Remote Sensing, 2025

    2025

  50. [50]

    Efficiently modeling long sequences with structured state spaces,

    A. Gu, K. Goel, and C. R ´e, “Efficiently modeling long sequences with structured state spaces,”arXiv preprint arXiv:2111.00396, 2021

    Pith/arXiv arXiv 2021

  51. [51]

    On the parameterization and initialization of diagonal state space models,

    A. Gu, K. Goel, A. Gupta, and C. R ´e, “On the parameterization and initialization of diagonal state space models,”Advances in Neural Information Processing Systems, vol. 35, pp. 35 971–35 983, 2022

    2022

  52. [52]

    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

    Pith/arXiv arXiv 2023

  53. [53]

    Videomamba: State space model for efficient video understanding,

    K. Li, X. Li, Y . Wang, Y . He, Y . Wang, L. Wang, and Y . Qiao, “Videomamba: State space model for efficient video understanding,” in European Conference on Computer Vision, 2024, pp. 237–255

    2024

  54. [54]

    Localmamba: Visual state space model with windowed selective scan,

    T. Huang, X. Pei, S. You, F. Wang, C. Qian, and C. Xu, “Localmamba: Visual state space model with windowed selective scan,” inEuropean Conference on Computer Vision. Springer, 2024, pp. 12–22

    2024

  55. [55]

    Vmamba: Visual state space model,

    Y . Liu, Y . Tian, Y . Zhao, H. Yu, L. Xie, Y . Wang, Q. Ye, J. Jiao, and Y . Liu, “Vmamba: Visual state space model,”Advances in Neural Information Processing Systems, vol. 37, pp. 103 031–103 063, 2024

    2024

  56. [56]

    Spectral-temporal token-guided prompt mamba for hyperspectral object tracking,

    H. Wang, Y . Li, and W. Li, “Spectral-temporal token-guided prompt mamba for hyperspectral object tracking,” in2024 14th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS), 2024, pp. 1–5

    2024

  57. [57]

    Bert: Pre-training of deep bidirectional transformers for language understanding,

    J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “Bert: Pre-training of deep bidirectional transformers for language understanding,” inPro- ceedings of NAACL-HLT, pp. 4171–4186

  58. [58]

    Clustering by fast search and find of density peaks,

    A. Rodriguez and A. Laio, “Clustering by fast search and find of density peaks,”science, vol. 344, no. 6191, pp. 1492–1496, 2014

    2014

  59. [59]

    All in one: Exploring unified vision-language tracking with multi-modal alignment,

    C. Zhang, X. Sun, Y . Yang, L. Liu, Q. Liu, X. Zhou, and Y . Wang, “All in one: Exploring unified vision-language tracking with multi-modal alignment,” inProceedings of the 31st ACM International Conference on Multimedia, 2023, pp. 5552–5561

    2023

  60. [60]

    Joint feature learning and relation modeling for tracking: A one-stream framework,

    B. Ye, H. Chang, B. Ma, S. Shan, and X. Chen, “Joint feature learning and relation modeling for tracking: A one-stream framework,” inEuropean Conference on Computer Vision. Springer, 2022, pp. 341– 357

    2022

  61. [61]

    Lora: Low-rank adaptation of large language models

    E. J. Hu, Y . Shen, P. Wallis, Z. Allen-Zhu, Y . Li, S. Wang, L. Wang, W. Chenet al., “Lora: Low-rank adaptation of large language models.” ICLR, vol. 1, no. 2, p. 3, 2022

    2022

  62. [62]

    Phtrack: Prompting for hyperspectral video tracking,

    Y . Chen, Y . Tang, X. Su, J. Li, Y . Xiao, J. He, and Q. Yuan, “Phtrack: Prompting for hyperspectral video tracking,”IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1–18, 2024. 13 ROYet al.: VISION-LANGUAGE GUIDED HYPERSPECTRAL OBJECT TRACKING XXXX

    2024

  63. [63]

    Sense: Hyperspectral video object tracker via fusing material and motion cues,

    Y . Chen, Q. Yuan, Y . Tang, Y . Xiao, J. He, and Z. Liu, “Sense: Hyperspectral video object tracker via fusing material and motion cues,” Information Fusion, vol. 109, p. 102395, 2024

    2024

  64. [64]

    Domain adaptation- aware transformer for hyperspectral object tracking,

    Y . Wu, L. Jiao, X. Liu, F. Liu, S. Yang, and L. Li, “Domain adaptation- aware transformer for hyperspectral object tracking,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 34, no. 9, pp. 8041– 8052, 2024

    2024

  65. [65]

    Ubstrack: Unified band selection and multimodel ensemble for hyperspectral object tracking,

    M. A. Islam, J. Zhou, W. Xing, Y . Gao, and K. K. Paliwal, “Ubstrack: Unified band selection and multimodel ensemble for hyperspectral object tracking,”IEEE Transactions on Geoscience and Remote Sensing, vol. 63, pp. 1–15, 2025

    2025

  66. [66]

    Causal hyperprompter: A framework for unbiased hyperspectral camouflaged object tracking,

    H. Wang, W. Li, X.-G. Xia, and Q. Du, “Causal hyperprompter: A framework for unbiased hyperspectral camouflaged object tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 1–18, 2025

    2025

  67. [67]

    Autoregressive queries for adaptive tracking with spatio-temporal trans- formers,

    J. Xie, B. Zhong, Z. Mo, S. Zhang, L. Shi, S. Song, and R. Ji, “Autoregressive queries for adaptive tracking with spatio-temporal trans- formers,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 19 300–19 309

    2024

  68. [68]

    Explicit visual prompts for visual object tracking,

    L. Shi, B. Zhong, Q. Liang, N. Li, S. Zhang, and X. Li, “Explicit visual prompts for visual object tracking,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 5, 2024, pp. 4838– 4846

    2024

  69. [69]

    Exploring enhanced contextual information for video-level object tracking,

    B. Kang, X. Chen, S. Lai, Y . Liu, Y . Liu, and D. Wang, “Exploring enhanced contextual information for video-level object tracking,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 39, no. 4, 2025, pp. 4194–4202

    2025

  70. [70]

    Hyper- track: A unified network for hyperspectral video object tracking,

    Y . Tan, W. Sun, J. Li, S. Hou, X. Li, Z. Wang, and B. Song, “Hyper- track: A unified network for hyperspectral video object tracking,”IEEE Transactions on Circuits and Systems for Video Technology, vol. 36, no. 1, pp. 1015–1028, 2026

    2026

  71. [71]

    Suit: Spatial-spectral union-intersection interaction network for hyperspectral object tracking,

    F. Xiong, Z. Wu, J. Zhou, S. Jia, and Y . Qian, “Suit: Spatial-spectral union-intersection interaction network for hyperspectral object tracking,” IEEE Transactions on Image Processing, vol. 34, pp. 7786–7800, 2025

    2025

  72. [72]

    Squeeze-and-excitation networks,

    J. Hu, L. Shen, and G. Sun, “Squeeze-and-excitation networks,” in2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018, pp. 7132–7141

    2018

  73. [73]

    Hyperspectral imagery band selection based on maximal standard deviation,

    L. Zhao, L. Wang, and D. Liu, “Hyperspectral imagery band selection based on maximal standard deviation,” in2015 8th International Sympo- sium on Computational Intelligence and Design (ISCID), vol. 2, 2015, pp. 59–62. 14

    2015