Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

MrTrack: Register Mamba for Needle Tracking with Rapid Reciprocating Motion during Ultrasound-Guided Aspiration Biopsy

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read MrTrack is proposed as the first ultrasound needle tracker for rapid reciprocating aspiration motion: a Mamba-based register extractor stores compact per-frame contexts, and a register retriever folds them into the template when current…

desk verdict Solid empirical tracker for a real clinical gap; the missing optical-to-ultrasound calibration protocol is a load-bearing gap in the paper's central claim. read the letter →

arxiv 2505.09450 v2 pith:HY3M2FJU submitted 2025-05-14 cs.CV

Yuelin Zhang , Qingpeng Ding , Long Lei , Yongxuan Feng , Raymond Shing-Yan Tang , Shing Shin Cheng This is my paper · ORCID
classification cs.CV
keywords ultrasound-guidedfineneedleaspirationtiptrackingMambastatespacemodelstemporalcontextregistertokensvisualobjectdiversifyloss
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

Ultrasound-guided fine-needle aspiration deliberately jabs the needle back and forth, which makes its tip blur or vanish between frames, and no existing needle tracker was designed for that motion. The paper proposes MrTrack, a tracker built around a Mamba-based register mechanism: each historical search frame is condensed into a compact register token, stored in a bank of the last 300 frames, and later retrieved to supplement the template when the current image is degraded. A self-supervised register diversify loss keeps these stored tokens from collapsing into identical features. On robotic and manual aspiration datasets, the paper reports that MrTrack is more accurate and more robust than seven comparison trackers while running at 73.9 frames per second, and that it generalizes to manual insertions it was never trained on.

What carries the argument

The central mechanism is a learnable register token paired with two Mamba (state-space sequence model) modules. In the extractor, the register is interleaved evenly behind each image segment of the search map before the selective-scan SSM, which condenses that frame's global appearance into $k=8$ compact descriptors; these are stored in the register bank, keeping the last $L=300$ frames. In the retriever, the historical descriptors are inserted before each segment of the template map and processed by another SSM, so temporal cues from the bank are melted into a dynamic template. The register diversify loss is the regularization that prevents collapse: it applies a softplus penalty to dimensions whose variance falls below $\tau=1$ and pushes cross-register covariances toward zero across different frames, encouraging dimension independence within the register bank.

What would settle it

Measure the reprojection error of a known 3D point from the optical tracker into the ultrasound image using a calibrated probe geometry; if this calibration error is comparable to or larger than MrTrack's reported mean errors (2.29 mm robotic, 2.80 mm manual), the claimed accuracy advantage would not be distinguishable from registration noise.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that temporal context in needle tracking can be stored frame-by-frame as compact registers and retrieved only when needed, instead of being propagated through the network as a running token. The Mamba-based register extractor inserts a learnable register of $k=8$ tokens into the sequence of image tokens before a selective-scan state-space model, so each frame's global context is distilled into descriptors that are appended to a register bank of length $L=300$. The Mamba-based register retriever performs the reverse insertion, placing historical descriptors before segments of the template sequence, producing a dynamic template that combines the initial template with cues from the last 300 frames. The register diversify loss combines a variance-promotion term and a cross-register covariance decorrelation term, both scaled by $0.01$, to keep the registers informative. The empirical claim is that this design outperforms the comparison trackers on AUC, precision, mean error and error standard deviation for robotic and manual aspiration biopsy, with the largest relative gains in mean error and standard deviation.

Load-bearing premise

The load-bearing premise is that the optical tracker's 3D ground-truth needle tip positions are accurately registered into the 2D ultrasound image coordinates, and the paper does not describe or verify that registration; if the mapping is wrong, all reported millimeter-level errors inherit that error.

Editorial extensions

If this is right

  • If the reported results hold, MrTrack would be the first needle tracker capable of following the rapid reciprocating motion of fine-needle aspiration, a motion regime that breaks trackers designed for stable insertions.
  • The per-frame register storage is presented as a cure for error accumulation in temporal token propagation: replacing the Mamba-based extractor and retriever with a transformer variant raises robotic mean error by $0.81$ mm, and removing the register entirely raises it by $0.20$ mm.
  • The tracker runs at 73.9 FPS on the reported hardware with a $384\times384$ search map and $192\times192$ template, fast enough for real-time use in high-frame-rate ultrasound.
  • The register diversify loss contributes a measurable gain: without it, robotic mean AUC drops 1.3 points and manual AUC drops 2.6 points, supporting the claim that decorrelated registers carry more useful temporal information.
  • The method generalizes without training to manual insertions with two to three needle oscillations per second, since the manual dataset was used only for testing.

Reading between the lines

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

  • Because the mechanism only depends on a history of search-image features, the same extractor-retriever design could be carried over to other interventional ultrasound tasks where a target periodically disappears, such as catheter or guidewire tracking; the paper does not test this.
  • The reported millimeter-level accuracies rest on an unverified projection of the optical tracker's 3D ground truth into ultrasound image coordinates, so the fairest next experiment is a full calibration-error analysis rather than further tracking-metric comparisons.
  • The ablation with an unbounded register bank ($L=\infty$) slightly improves robotic accuracy but hurts manual accuracy and drops speed by 15.9 FPS, which suggests that a learned forgetting or importance-weighting mechanism over the register bank could beat the fixed $L=300$ window.
  • For clinical translation, training on manual and real clinical video may shift the optimal bank length and register size, since manual insertions contain more abrupt imaging disturbances than the robotic dataset.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper proposes MrTrack, a Mamba-based visual tracker with learnable registers for needle-tip tracking in ultrasound-guided fine needle aspiration (FNA) biopsy. A Mamba-based register extractor condenses each historical search map into compact register tokens stored in a register bank; a Mamba-based register retriever injects these temporal prompts into the template to handle rapid reciprocating motion and transient image degradation. A self-supervised register diversify (RD) loss is introduced to prevent feature collapse. Experiments on a robotic dataset (239 trials, 54,428 frames) and a manual dataset (16 videos used only for testing) compare MrTrack against seven baselines, showing lower localization error, higher AUC/precision, and higher inference speed (73.9 FPS). Ablations examine register-bank length, register scale, transformer-based alternatives, removal of the register mechanism, and removal of the RD loss.

Significance. If the quantitative claims hold, MrTrack would be the first tracker explicitly designed for the rapid reciprocating needle motion of FNA biopsy, and its efficiency advantage is practically relevant for real-time ultrasound-guided intervention. The paper provides a broad comparison with seven baselines under shared training settings, an ablation study of the main components, and generalization testing on a manual dataset that was not used for training. However, the evaluation protocol has two load-bearing gaps: the optical-tracker-to-ultrasound coordinate mapping is not described, and no run-to-run or trial-level statistical variability is reported. The RD-loss equation also has an ambiguity that prevents exact reproduction. These issues are fixable and should be addressed before the central accuracy claims can be accepted.

major comments (3)
  1. [Section 4.1 / Table 1] The ground-truth mapping from optical-tracker 3D coordinates to 2D ultrasound image coordinates is not described. Section 4.1 reports only that the ClaroNav MicronTracker 3 has an RMSE of 0.189 mm in the authors' setup; it does not specify the calibration procedure, the rigid transformation between tracker and ultrasound coordinate frames, the sonographic pixel-to-millimeter scaling, or the temporal synchronization between the optical tracker and the 30 FPS ultrasound frames. All reported metrics (Err, AUC, P) compare tracker outputs against this 2D ground truth, so a biased or mis-synchronized projection directly affects every quantitative conclusion. At a needle speed of 30 mm/s, a one-frame temporal misalignment shifts the ground truth by about 1 mm, comparable to the reported mean errors. The authors should provide the calibration protocol, reprojection error, temporal synchronization method, and a sensitivity analysis of the main metrics to plausible calibration errors.
  2. [Table 1 / Section 4.2] Only frame-level standard deviation of the localization error is reported; no run-to-run variance, number of repeated training runs, random seeds, or trial-level confidence intervals are given. The headline improvements over the second-best method (21.8% Err reduction on the robotic dataset and 8.2% on the manual dataset) could be within training or trial variability. I ask the authors to report results over at least three training runs (or otherwise justify that the comparisons are stable), and to provide per-trial or per-video error statistics and appropriate significance tests for the manual dataset, which contains only 16 videos.
  3. [Equation (4), Section 3.4] The cross-register diversify term L_div is not well-defined as written. The text defines each descriptor r_i as an element of R^{k×d}, so r_{i,p} is a k-dimensional vector, and the expression [(r_{i,p} - \bar{r})(r_{j,p} - \bar{r})]^2 is ambiguous: it is unclear whether the product is an inner product, an outer product, or an elementwise operation, and whether the square is a scalar or a matrix norm. Since L_div is a core component of the proposed RD loss, this equation must be rewritten with the exact tensor operations and the resulting scalar form, and the implementation should be clarified to match.
minor comments (5)
  1. [Section 2] There is a typo in the Related Work section: 'learning-based methods have bee proposed' should read 'have been proposed'.
  2. [Abstract and Section 1] The name 'MrT rack' appears with an unusual space in several places (e.g., 'MrT rack, an ultrasound needle tracker'); this should be normalized to 'MrTrack' throughout.
  3. [Section 4.3] The sentence 'The robustness of L_RD is also showed by avoiding tuning the value of hyper-parameters α and β' is not supported by any sensitivity analysis; the paper fixes α=β=0.01 but does not show that nearby values give similar performance. Either add such an analysis or rephrase the claim.
  4. [Table 1] The table caption does not define IPS/IPM; the terms are explained in Section 4.1 but should also be defined in the caption for readability.
  5. [Section 4.1] The paper does not state whether the same optical-tracker-to-ultrasound calibration was used for both the robotic and manual datasets; please clarify whether the manual dataset uses the same geometric setup and synchronization, since this affects the interpretation of the generalization results.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: MrTrack's contributions are empirical architectural components evaluated on held-out test data, with no prediction that reduces to a fitted input by construction.

full rationale

MrTrack is an empirical method paper. The claimed contributions are a Mamba-based register extractor, a register retriever, and a self-supervised register diversify loss. None of these is derived from or defined in terms of the reported accuracy metrics (Err, AUC, P). The RD loss (Eq. 3-5) is a regularizer applied during training; it does not produce the evaluation numbers by construction, and its hyperparameters are fixed scalar weights rather than fitted predictions. The ablation study compares the full model against variants with components removed, which is a standard empirical attribution and does not presuppose the conclusion. The self-citations [33,34,35] appear only in related-work context, and the central register concept is cited to external prior work [4,25]. No uniqueness theorem or load-bearing argument is imported from the authors' own prior papers. The reader's concern about the undocumented 3D-to-2D ground-truth registration is a validation/reproducibility issue affecting correctness, not a circularity of the derivation chain, because the paper's equations and method do not use that calibration as an input that is then renamed as a prediction. The comparison methods are trained under the same protocol, and the manual dataset is held out, so the generalization claim is externally checkable in principle. Thus the paper's central claim is self-contained with respect to circularity.

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

The central claim rests on standard deep-learning assumptions and the unverified optical-to-ultrasound registration. Hyperparameters L, k, alpha, beta, tau are hand-chosen.

free parameters (4)
  • register bank length L = 300
    Set to 300 in the baseline; ablation vr1/vr2 shows sensitivity, with L=300 chosen as moderate.
  • register scale k = 8
    Number of register tokens per frame; ablation vr3/vr4 shows k=8 is best.
  • RD loss weights alpha and beta = 0.01 each
    Chosen scalar weights for variance and decorrelation terms; paper states robustness to tuning is shown.
  • variance threshold tau = 1
    Threshold in L_var to promote variance; set to 1.
assumptions (4)
  • domain assumption The optical tracker ground truth can be accurately registered to the 2D ultrasound image.
    Section 4.1 only gives the RMSE of the optical tracker, but not the calibration between the tracker and the US image. The entire evaluation depends on this mapping being accurate.
  • domain assumption The needle tip remains visible or trackable in the ultrasound image during aspiration.
    The method assumes temporal context can compensate for temporary occlusion or artifacts; if the tip is entirely invisible, no tracker could work.
  • domain assumption Pre-trained ViT and Mamba provide suitable features for ultrasound needle tracking.
    The method relies on a ViT-Base backbone and Mamba's SSM to extract features, assuming these generalize to ultrasound images.
  • domain assumption The comparative trackers were trained fairly with the same data and settings.
    The paper states all models were trained with the same strategy, but no code or hyperparameter details are provided for baselines, so this is an assumption.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MrTrack: Register Mamba for Needle Tracking with Rapid Reciprocating Motion during Ultrasound-Guided Aspiration Biopsy." pith.science (2026). https://pith.science/paper/HY3M2FJU

@misc{pith2026250509450,
  author       = {Pith},
  title        = {Pith review of: MrTrack: Register Mamba for Needle Tracking with Rapid Reciprocating Motion during Ultrasound-Guided Aspiration Biopsy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HY3M2FJU}},
  note         = {Machine review of arXiv:2505.09450}
}
read the original abstract

Ultrasound-guided fine needle aspiration (FNA) biopsy is a common minimally invasive diagnostic procedure. However, an aspiration needle tracker addressing rapid reciprocating motion is still missing. MrTrack, an aspiration needle tracker with a mamba-based register mechanism, is proposed. MrTrack leverages a Mamba-based register extractor to sequentially distill global context from each historical search map, storing these temporal cues in a register bank. The Mamba-based register retriever then retrieves temporal prompts from the register bank to provide external cues when current vision features are temporarily unusable due to rapid reciprocating motion and imaging degradation. A self-supervised register diversify loss is proposed to encourage feature diversity and dimension independence within the learned register, mitigating feature collapse. Comprehensive experiments conducted on both robotic and manual aspiration biopsy datasets demonstrate that MrTrack not only outperforms state-of-the-art trackers in accuracy and robustness but also achieves superior inference efficiency. Project page: https://github.com/PieceZhang/MrTrack

Figures

Figures reproduced from arXiv: 2505.09450 by the authors.

Figure 1
Figure 1. Overview of MrTrack. The notes (e.g. 3 × 3, s = 1) on convolution layers represent the kernel size and stride. 3.2 Overview of MrTrack As shown in [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Experiment setup (a) and tracking demonstration (b). More demonstra [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Real-Time Guidewire Tip Tracking Using a Siamese Network for Image-Guided Endovascular Procedures

    eess.IV 2025-06 conditional novelty 4.0 of 10

    A Siamese network with edge-enhanced attention tracks guidewire tips in DSA video at 57 FPS, with 0.421 mm mean error on three test sequences and 0.148-0.708 mm error on two robotic tasks.

Reference graph

Works this paper leans on

36 extracted references · 27 canonical work pages · cited by 1 Pith paper

  1. [1]

    arXiv preprint arXiv:2105.04906 (2021)

    Bardes, A., et al.: Vicreg: Variance-invariance-covariance regularization for self- supervised learning. arXiv preprint arXiv:2105.04906 (2021)

    arXiv 2021

  2. [2]

    IEEE Robotics and Automation Letters (2024)

    Cai, Y., et al.: Autonomous robotic ultrasound approach for fetoscope tracking by fusing optical and 2d ultrasound data. IEEE Robotics and Automation Letters (2024)

    work page 2024

  3. [3]

    IEEE Transactions on Pattern Analysis and Machine Intelligence45(4), 5158–5173 (2022)

    Chen, Z., et al.: Siamban: Target-aware tracking with siamese box adaptive net- work. IEEE Transactions on Pattern Analysis and Machine Intelligence45(4), 5158–5173 (2022)

    work page 2022

  4. [4]

    arXiv preprint arXiv:2309.16588 (2023)

    Darcet, T., et al.: Vision transformers need registers. arXiv preprint arXiv:2309.16588 (2023)

    arXiv 2023

  5. [5]

    arXiv preprint arXiv:2010.11929 (2020)

    Dosovitskiy, A.: An image is worth 16x16 words: Transformers for image recogni- tion at scale. arXiv preprint arXiv:2010.11929 (2020)

    arXiv 2020

  6. [6]

    European radiology26, 1801–1807 (2016)

    D’Onofrio, M., et al.: Ultrasound-guided percutaneous fine-needle aspiration of solid pancreatic neoplasms: 10-year experience with more than 2,000 cases and a review of the literature. European radiology26, 1801–1807 (2016)

    work page 2016

  7. [7]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Fu, Z., et al.: Stmtrack: Template-free visual tracking with space-time memory networks. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 13774–13783 (2021)

    work page 2021

  8. [8]

    Advances in neural information processing systems33, 21271–21284 (2020)

    Grill, J.B., et al.: Bootstrap your own latent-a new approach to self-supervised learning. Advances in neural information processing systems33, 21271–21284 (2020)

    work page 2020

Show all 36 references
  1. [9]

    arXiv preprint arXiv:2312.00752 (2023)

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

    2023 arXiv

  2. [10]

    IEEE Transactions on Pattern Analysis and Machine Intelligence (2024) 10 Y

    Gui, J., et al.: A survey on self-supervised learning: Algorithms, applications, and future trends. IEEE Transactions on Pattern Analysis and Machine Intelligence (2024) 10 Y. Zhang et al

    2024

  3. [11]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Guo, D., et al.: Siamcar: Siamese fully convolutional classification and regression for visual tracking. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 6269–6277 (2020)

    2020

  4. [12]

    Kalman, R.E.: A new approach to linear filtering and prediction problems (1960)

    1960

  5. [13]

    Artificial Intelligence Surgery 4(3), 149–169 (2024)

    Kimbowa, A., et al.: Advancements in needle visualization enhancement and local- ization methods in ultrasound: a literature review. Artificial Intelligence Surgery 4(3), 149–169 (2024)

    2024

  6. [14]

    arXiv preprint arXiv:2408.07889 (2024)

    Lai, S., et al.: Mambavt: Spatio-temporal contextual modeling for robust rgb-t tracking. arXiv preprint arXiv:2408.07889 (2024)

    2024 arXiv

  7. [15]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Li, B., et al.: Siamrpn++: Evolution of siamese visual tracking with very deep networks. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 4282–4291 (2019)

    2019

  8. [16]

    Advances in Neural Information Processing Systems35, 16743–16754 (2022)

    Lin, L., et al.: Swintrack: A simple and strong baseline for transformer tracking. Advances in Neural Information Processing Systems35, 16743–16754 (2022)

    2022

  9. [17]

    Medical Image Analysis87, 102810 (2023)

    Luo, M., et al.: Recon: Online learning for sensorless freehand 3d ultrasound re- construction. Medical Image Analysis87, 102810 (2023)

    2023

  10. [18]

    In: 2024 International Symposium on Medical Robotics (ISMR)

    Ma, X., et al.: Feasibility of pointcloud-based ultrasound-ct registration towards automated, robot-assisted image-guidance in spine surgery. In: 2024 International Symposium on Medical Robotics (ISMR). pp. 1–7. IEEE (2024)

    2024

  11. [19]

    Journal of Medical Robotics Research p

    Ma, X., et al.: Cross-modality registration using bone surface pointcloud for robotic ultrasound-guided spine surgery. Journal of Medical Robotics Research p. 2540004 (2025)

    2025

  12. [20]

    In: Proceedings of the European conference on computer vision (ECCV)

    Muller, M., et al.: Trackingnet: A large-scale dataset and benchmark for object tracking in the wild. In: Proceedings of the European conference on computer vision (ECCV). pp. 300–317 (2018)

    2018

  13. [21]

    In: Inter- national Conference on Medical Image Computing and Computer-Assisted Inter- vention

    Mwikirize, C., et al.: Single shot needle tip localization in 2d ultrasound. In: Inter- national Conference on Medical Image Computing and Computer-Assisted Inter- vention. pp. 637–645. Springer (2019)

    2019

  14. [22]

    Atlas of Acquired Cardiovascular Disease Imaging in Children pp

    Richardson, MD, R.R., Richardson, R.R.: Imaging modalities: Advantages and dis- advantages. Atlas of Acquired Cardiovascular Disease Imaging in Children pp. 1–4 (2017)

    2017

  15. [23]

    Elsevier Health Sciences (2017)

    Rumack, C.M., Levine, D.: Diagnostic ultrasound. Elsevier Health Sciences (2017)

    2017

  16. [24]

    Shen, C., et al.: Discriminative correlation filter network for robust landmark track- inginultrasoundguidedintervention.In:MedicalImageComputingandComputer Assisted Intervention–MICCAI 2019: 22nd International Conference, Shenzhen, China, October 13–17, 2019, Proceedings, Par...

    2019

  17. [25]

    arXiv preprint arXiv:2405.14858 (2024)

    Wang, F., et al.: Mamba-r: Vision mamba also needs registers. arXiv preprint arXiv:2405.14858 (2024)

    2024 arXiv

  18. [26]

    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sens- ing (2024)

    Wang, Q., et al.: Trackingmamba: Visual state space model for object tracking. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sens- ing (2024)

    2024

  19. [27]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Wu, Y., et al.: Online object tracking: A benchmark. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 2411–2418 (2013)

    2013

  20. [28]

    arXiv preprint arXiv:2412.13611 (2024)

    Xie, J., et al.: Robust tracking via mamba-based context-aware token learning. arXiv preprint arXiv:2412.13611 (2024)

    2024 arXiv

  21. [29]

    Medical Image Analysis88, 102847 (2023)

    Yan, W., et al.: Learning-based needle tip tracking in 2d ultrasound by fusing visual tracking and motion prediction. Medical Image Analysis88, 102847 (2023)

    2023

  22. [30]

    IEEE Transactions on Medical Imaging (2024) MrTrack: Needle Tracking during Aspiration Biopsy 11

    Yan, W., et al.: Task-oriented network design for visual tracking and motion fil- tering of needle tip under 2d ultrasound. IEEE Transactions on Medical Imaging (2024) MrTrack: Needle Tracking during Aspiration Biopsy 11

    2024

  23. [31]

    In: International conference on machine learning

    Zbontar,J.,etal.:Barlowtwins:Self-supervisedlearningviaredundancyreduction. In: International conference on machine learning. pp. 12310–12320. PMLR (2021)

    2021

  24. [32]

    Applied Sciences14(13), 5683 (2024)

    Zhang, H., et al.: A survey on visual mamba. Applied Sciences14(13), 5683 (2024)

    2024

  25. [33]

    arXiv preprint arXiv:2403.05146 (2024)

    Zhang, Y., et al.: Motion-guided dual-camera tracker for endoscope tracking and motion analysis in a mechanical gastric simulator. arXiv preprint arXiv:2403.05146 (2024)

    2024 arXiv

  26. [34]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Zhang, Y., et al.: A unified framework for microscopy defocus deblur with multi- pyramid transformer and contrastive learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11125–11136 (2024)

    2024

  27. [35]

    Zhang, Y., et al.: Mambaxctrack: Mamba-based tracker with ssm cross-correlation andmotionpromptforultrasoundneedletracking.IEEERoboticsandAutomation Letters (2025)

    2025

  28. [36]

    In:ProceedingsoftheAAAIConferenceonArtificialIntelligence.vol.38,pp.7588– 7596 (2024)

    Zheng,Y.,etal.:Odtrack:Onlinedensetemporaltokenlearningforvisualtracking. In:ProceedingsoftheAAAIConferenceonArtificialIntelligence.vol.38,pp.7588– 7596 (2024)

    2024

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.