Pith. sign in

REVIEW 3 major objections 4 minor 51 references

Low-Contrast-Enhanced Contrastive Learning for Semi-Supervised Endoscopic Image Segmentation

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read The paper claims that deliberately mining low-contrast pixels for contrastive learning is what pushes semi-supervised endoscopic segmentation past previous methods, especially for minority tumor classes.

desk verdict LoCo is a sensible, incremental combination of density-guided contrastive learning and adaptive pseudo-label filtering for semi-supervised endoscopic segmentation, but the SOTA claim outruns the statistics: no error bars anywhere and one public-dataset cell actually trails AugSeg. read the letter →

arxiv 2412.02314 v2 pith:KAPXASUM submitted 2024-12-03 cs.CV

classification cs.CV
keywords semi-supervisedsemanticsegmentationcontrastivelearningendoscopicimagelow-contrastpixelspseudo-labelfilteringmean-teacherminorityclassmedical
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

Semi-supervised endoscopic image segmentation struggles when the boundary between tumor and normal tissue is faint and when benign and malignant tumors look alike. This paper argues that the pixels the model finds hardest to classify—low-contrast pixels—should be the primary source of contrastive supervision, not all pixels equally. It proposes LoCo, a mean-teacher framework whose low-contrast-enhanced contrastive loss (LCC) is built from two selection rules: inter-class contrast enhancement (ICE) picks pixels whose features sit farthest from their class average, and boundary contrast enhancement (BCE) picks boundary pixels whose features are most similar to their neighbours. A confidence-based dynamic filter (CDF) then adjusts pseudo-label thresholds per class so rare classes such as benign tumors contribute more unlabeled supervision. On a proprietary laryngeal cancer dataset and two public polyp datasets, LoCo reports the highest mIoU among the compared methods at 10%, 30%, and 50% labeled partitions, with the largest margins on the minority benign class.

What carries the argument

The load-bearing machinery is the low-contrast-enhanced contrastive loss (LCC), a per-pixel contrastive objective whose samples are chosen by two selection rules rather than drawn uniformly. Inter-class contrast enhancement (ICE) uses class embeddings $\bar{z}_c$ obtained by class-wise average pooling of labeled features (Eq. 9) and keeps the top $k\%$ of pixels with lowest cosine similarity to their own class embedding (Eq. 12). Boundary contrast enhancement (BCE) defines boundary pixels by a neighbourhood label check and keeps the top $k\%$ with highest boundary feature similarity (Eq. 13). The contrastive loss (Eq. 11) then pulls each selected embedding toward its class embedding and pushes it away from the other class embeddings, and the confidence-based dynamic filter (CDF) sets per-class pseudo-label thresholds through EMA-updated global and local confidence estimates (Eqs. 14–18). This selection-plus-contrast mechanism is what the ablation study credits for most of the reported improvement over the unlabeled-supervision baseline.

What would settle it

Run LoCo on the laryngeal cancer dataset under a 10% label partition but replace the ICE pixel-selection rule with random pixel selection at the same $k\%$, keeping BCE and CDF fixed; the paper's claim predicts a large drop in benign-class IoU relative to the full model, while a noise-sensitivity alternative predicts little change. A second check is to compare the selected low-similarity pixels against expert-annotated low-contrast regions: if the selection is doing its job, most selected pixels should fall inside or immediately beside regions experts mark as low contrast.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that explicitly selecting low-contrast pixels for contrastive learning yields a large and consistent gain in semi-supervised endoscopic segmentation. ICE computes a class embedding $\bar{z}_c$ for each class by average-pooling labeled feature embeddings, then selects the top $k\%$ of pixels with the lowest cosine similarity to their own class embedding; BCE selects boundary pixels with the highest minimum similarity to their $h$ neighbours. The low-contrast-enhanced contrastive loss (Eq. 11) pulls those selected embeddings toward the correct class embedding and pushes them away from the other class embeddings. CDF replaces a fixed pseudo-label threshold with an EMA-updated global threshold scaled by per-class local thresholds, increasing the number of pseudo-labels kept for minority classes. Together the components improve mIoU on the laryngeal cancer dataset by 2.55, 3.41, and 2.75 points over the previous best method at 10%, 30%, and 50% labels, and the benign-class IoU by 4.13, 5.29, and 3.58 points; the full model also reports the best mIoU on both polyp datasets.

Load-bearing premise

The method assumes that the pixels whose learned features sit farthest from their class's average feature are genuinely hard low-contrast pixels worth extra training, rather than noise or mislabeled examples—an assumption that becomes fragile when only 10% of images are labeled and those class averages come from very few images.

Editorial extensions

If this is right

  • Under a 10% label partition on the laryngeal cancer dataset, LoCo reports a benign-tumor IoU of 52.96%, a 4.13-point gain over the previous best method, showing that minority-class pixels benefit the most from hard-pixel contrastive learning.
  • On Kvasir-SEG, LoCo reports mIoU gains over the second-best method of 2.57, 1.61, and 1.83 points at 10%, 30%, and 50% labels, indicating the benefit persists when labels are more plentiful.
  • The ablation study (M5 vs M2) attributes the largest single contribution to ICE rather than BCE, suggesting that inter-class hard-pixel mining is the more critical of the two selection rules on this data.
  • CDF raises pseudo-label utilization for the minority benign class relative to a fixed 0.95 threshold, implying that fewer rare-class labels are discarded during unlabeled training.
  • The authors position LoCo as applicable beyond laryngeal cancer to polyp segmentation, where the same framework reports the best mIoU on both public datasets.

Reading between the lines

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

  • The ICE selection rule implicitly assumes each class has a single compact prototype in embedding space; if a lesion class is visually heterogeneous, the lowest-similarity pixels may be legitimate subclusters rather than hard negatives, and pulling them to one prototype could distort the feature space. Testing class-conditional clustering before selection would clarify this.
  • Because CDF's thresholds are driven by teacher confidence, early-training confidences could let noisy pseudo-labels through for minority classes before the model stabilises; a confidence-calibration or warm-up schedule is a natural extension the paper does not explore.
  • The same hard-pixel selection idea should transfer to other low-contrast dense prediction tasks such as skin-lesion or surgical-scene segmentation, but this is an extrapolation: the paper only evaluates endoscopy.
  • If the gain comes from selection rather than contrastive loss per se, then replacing ICE's low-similarity pixels with randomly chosen pixels at the same $k\%$ should erase most of the reported improvement, a comparison the paper does not run.
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 / 4 minor

Summary. The paper proposes LoCo, a mean-teacher framework for semi-supervised endoscopic image segmentation. The method combines a supervised cross-entropy loss with an unsupervised consistency loss on pseudo-labels filtered by a confidence-based dynamic filter (CDF), and adds a low-contrast-enhanced contrastive loss (LCC) built from two selection strategies: inter-class contrast enhancement (ICE), which picks pixels with the lowest similarity to their class prototype, and boundary contrast enhancement (BCE), which picks boundary pixels with the highest local feature similarity. Experiments are reported on one proprietary laryngeal-cancer dataset (FAHSYU-LC) and two public polyp datasets (Kvasir-SEG, CVC-ClinicDB) under 10%, 30%, and 50% labeled-data partitions, against FixMatch, U2PL, UniMatch, AugSeg, and CorrMatch. The paper claims state-of-the-art performance, significantly outperforming previous methods, and provides an ablation study on FAHSYU-LC at 10% labels.

Significance. The problem addressed is relevant and timely: semi-supervised endoscopic segmentation with low-contrast boundaries and minority tumor classes is practically important, and the paper tackles an underexplored inter-class contrast issue. The proposed components are clearly motivated, the framework is simple to combine with existing mean-teacher pipelines, and the authors provide source code, which strengthens reproducibility. If the reported gains are reproducible, the method would be a meaningful advance, particularly for low-contrast and minority-class pixels. However, the evidence as presented is not sufficient to support the strength of the headline claim, because the comparisons rest on single runs and at least one reported cell contradicts the claim of consistent superiority. The largest gains are on a proprietary dataset, which limits external verification.

major comments (3)
  1. [IV-B, Tables I-III] The central claim that LoCo 'significantly outperforms previous methods' is not statistically supported. Tables I-III report a single run per setting with no error bars, no standard deviations, and no significance tests. Moreover, on CVC-ClinicDB at 10% labels (Table III), LoCo is below AugSeg on all three metrics: mIoU 83.81 vs. 84.13, DSC 76.31 vs. 78.49, and NSD 79.67 vs. 81.81; the text acknowledges this as 'on par', which is inconsistent with the abstract and conclusion. Because the margins elsewhere are 1-3 mIoU points, a paired test or multiple-seed reporting is needed to distinguish the claimed improvement from training noise. Please provide mean and standard deviation over at least three seeds and a paired significance test, and temper the claims accordingly.
  2. [III-B1, Eqs. (9) and (12)] The ICE selection rule relies on class embeddings computed by class-wise average pooling over the labeled batch (Eq. 9), and the batch size is 8 (Section IV-A3). Under a 10% label partition, the benign class may be represented by very few pixels per batch, making the prototype unstable and the top-k% lowest-similarity selection potentially driven by noise or mislabeled features. Since the ablation in Table IV attributes the largest gain to ICE (M5 vs. M2: mIoU 61.64 vs. 58.05), this stability concern is load-bearing. Please provide an analysis of prototype sensitivity, such as varying k, reporting performance across seeds, or showing that the selected pixels are consistently low-contrast rather than noise.
  3. [IV-C, Table IV] The ablation study is reported without variance, yet several component contributions are small and have mixed directions. For example, M4 (adding BCE to M2) increases mIoU by 0.65 but decreases IoU(M) from 70.42 to 69.91, and M3 (adding CDF to M2) decreases IoU(M) from 70.42 to 70.18. Without multiple runs, it is impossible to determine whether these differences are real or within noise. Please report standard deviations or significance tests for the ablation, and discuss which contributions are robust.
minor comments (4)
  1. [III-A, Eq. (3) and surrounding text] The description of weak perturbation says 'including random scale, random scale, and random flip'; the duplicate 'random scale' appears to be a typo, and the intended augmentation list should be stated precisely.
  2. [III-C, Eqs. (14) and (17)] In Eq. (14), the confidence term max(ˆy^u_{i,j}) uses the student prediction on the strongly augmented image, while the mask 1[yp_{i,j}=c] is derived from the teacher pseudo-label. Please clarify whether the confidence is intentionally taken from the student output; if so, justify this choice, and if not, correct the equation to use the teacher prediction.
  3. [III-B, III-B1, III-B2] The definition of the low-contrast feature embedding set H is informal. It is not specified whether the top k% selection in ICE and BCE is performed per class, per image, or globally across the batch, nor whether the same k is applied independently to labeled and unlabeled features. Please clarify the exact selection procedure.
  4. [IV-C, Table IV] The ablation table would be easier to interpret if it also reported the supervised-only baseline with the same pseudo-label and contrastive components removed stepwise; currently M2 already includes Lu, and the contributions of the individual losses relative to M1 are conflated with the effect of unlabeled data.

Circularity Check

1 steps flagged · score 2.0 of 10

LoCo's external SOTA comparisons are independent; only the ICE 'effectiveness' visualization is self-referential by construction.

  1. self definitional [Section III-B Eq. (11) and Section IV-D Fig. 3]
    "The loss of LCC can be defined as: Llcc = − 1/K Σ_c 1/|{z_c}| Σ_{z_i∈{z_c}} log exp(z_i · ¯z_c/τ) / (exp(z_i · ¯z_c/τ) + Σ_{z_j ∈H\{z_c}} exp(z_j · ¯z_c/τ)). ... Fig. 3 presents a visualization of the effectiveness of ICE compared to a baseline (i.e., M1) over the course of training epochs. The plot shows the inter-class feature similarity, which measures the average similarity between the feature embeddings and the class embeddings of different categories."

    The LCC loss in Eq. (11) explicitly maximizes the log-ratio between self-class similarity exp(z_i·zbar_c/τ) and the summed cross-class similarities exp(z_j·zbar_c/τ) for z_j in H\{z_c}. Minimizing this loss therefore directly decreases the exact quantity that Fig. 3 reports as evidence of ICE's effectiveness. ICE selects the lowest-similarity pixels and then pulls or pushes them relative to the same class embeddings used to define the metric, so the observed downward trend is a restatement of the training objective rather than an independent validation. This does not propagate to the main mIoU comparisons, which are evaluated against external baselines.

full rationale

The central SOTA claim is evaluated against five published baselines on two public datasets plus a proprietary dataset; no fitted parameter is relabeled as a prediction, and no load-bearing self-citation or imported uniqueness theorem appears. CDF uses the model's own prediction confidence to set pseudo-label thresholds, which is self-referential but is a standard self-training mechanism and does not by itself force the reported segmentation margins. The only noticeable circular element is the ICE-effectiveness visualization, whose metric is identical to the LCC objective; this is a minor self-validating demonstration, not a derivation of the main results. Overall circularity is therefore low.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The method is empirical, so there is no mathematical derivation to audit. The listed assumptions are the load-bearing modeling choices: class prototypes from small labeled sets, confidence as a correctness proxy, and two geometric definitions of hard pixels. Hyperparameters k, h, gamma, tau, lambda_1, lambda_2 are hand-set with no sensitivity analysis. No invented entities.

free parameters (7)
  • lambda_1 = 0.5
    Weight of unsupervised pseudo-label loss in Eq. (10); chosen empirically, no sensitivity study.
  • lambda_2 = 0.1
    Weight of low-contrast contrastive loss in Eq. (10); chosen empirically, no sensitivity study.
  • k = 30%
    Top 30% of pixels selected by ICE and BCE as low-contrast pixels; set in Section IV-A.3 without sensitivity analysis.
  • h = 64
    Neighborhood size defining boundary pixels in BCE, Eq. (13); fixed without ablation.
  • gamma = 0.25
    Exponent controlling local threshold influence in CDF, Eq. (18); fixed without ablation.
  • tau = 0.1
    Temperature in contrastive loss, Eq. (11); fixed without sensitivity analysis.
  • T_g_init = 0.85
    Initial global threshold in CDF, Eq. (15); fixed without sensitivity analysis.
assumptions (5)
  • domain assumption Teacher network predictions under weak augmentation are reliable enough to serve as pseudo-labels.
    The consistency loss Eq. (6) depends on this; if the teacher is overconfident, CDF cannot prevent confirmation bias.
  • domain assumption Class embeddings computed by class-wise average pooling over labeled data are stable class prototypes.
    Eq. (9) supplies the prototypes for ICE and the contrastive loss; with 10% labels and class imbalance these averages can be noisy.
  • domain assumption Cosine similarity between a pixel embedding and its class embedding is a valid measure of low contrast.
    ICE selects pixels with lowest s_c in Eq. (12); if similarity is dominated by unrelated factors, ICE selects the wrong pixels.
  • domain assumption Boundary pixels are identified by label disagreement within an h-pixel neighborhood, and high boundary feature similarity indicates difficulty.
    BCE uses Eq. (13); this assumes local label maps and feature locality.
  • domain assumption Pseudo-label confidence correlates with correctness for every class.
    CDF thresholds by max confidence in Eqs. (14)-(18); miscalibrated confidence would hurt minority-class pseudo-labels.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Low-Contrast-Enhanced Contrastive Learning for Semi-Supervised Endoscopic Image Segmentation." pith.science (2026). https://pith.science/paper/KAPXASUM

@misc{pith2026241202314,
  author       = {Pith},
  title        = {Pith review of: Low-Contrast-Enhanced Contrastive Learning for Semi-Supervised Endoscopic Image Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KAPXASUM}},
  note         = {Machine review of arXiv:2412.02314}
}
read the original abstract

The segmentation of endoscopic images plays a vital role in computer-aided diagnosis and treatment. The advancements in deep learning have led to the employment of numerous models for endoscopic tumor segmentation, achieving promising segmentation performance. Despite recent advancements, precise segmentation remains challenging due to limited annotations and the issue of low contrast. To address these issues, we propose a novel semi-supervised segmentation framework termed LoCo via low-contrast-enhanced contrastive learning (LCC). This innovative approach effectively harnesses the vast amounts of unlabeled data available for endoscopic image segmentation, improving both accuracy and robustness in the segmentation process. Specifically, LCC incorporates two advanced strategies to enhance the distinctiveness of low-contrast pixels: inter-class contrast enhancement (ICE) and boundary contrast enhancement (BCE), enabling models to segment low-contrast pixels among malignant tumors, benign tumors, and normal tissues. Additionally, a confidence-based dynamic filter (CDF) is designed for pseudo-label selection, enhancing the utilization of generated pseudo-labels for unlabeled data with a specific focus on minority classes. Extensive experiments conducted on two public datasets, as well as a large proprietary dataset collected over three years, demonstrate that LoCo achieves state-of-the-art results, significantly outperforming previous methods. The source code of LoCo is available at the URL of \href{https://github.com/AnoK3111/LoCo}{https://github.com/AnoK3111/LoCo}.

Figures

Figures reproduced from arXiv: 2412.02314 by the authors.

Figure 1
Figure 1. Illustration of low-contrast pixels in endoscopic images. Rows [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overall architecture of LoCo. It is designed based on the mean-teacher framework, featuring two branches: a student network and a teacher network. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 4
Figure 4. Visualization of boundary captured by BCE: (a) Ground truth, (b) [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: Visualization of boundary feature similarity at different epochs: (a) [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Visualization of the effectiveness of CDF: (a) The threshold for [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 8
Figure 8. Figure 8: Visualization of segmentation results on FAHSYU-LC with [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

51 extracted references · 22 canonical work pages

  1. [1]

    Advanced endoscopy meets molecular diagnosis of cholangiocarcinoma,

    M. Cadamuro, A. Al-Taee, and T. A. Gonda, “Advanced endoscopy meets molecular diagnosis of cholangiocarcinoma,” Journal of Hepatol- ogy, vol. 78, no. 5, pp. 1063–1072, 2023

  2. [2]

    A review of deep learning-based multiple- lesion recognition from medical images: classification, detection and segmentation,

    H. Jiang, Z. Diao, T. Shi, Y . Zhou, F. Wang, W. Hu, X. Zhu, S. Luo, G. Tong, and Y .-D. Yao, “A review of deep learning-based multiple- lesion recognition from medical images: classification, detection and segmentation,” Computers in Biology and Medicine, vol. 157, p. 106726, 2023

  3. [3]

    Encoder- decoder with atrous separable convolution for semantic image segmen- tation,

    L.-C. Chen, Y . Zhu, G. Papandreou, F. Schroff, and H. Adam, “Encoder- decoder with atrous separable convolution for semantic image segmen- tation,” in Proceedings of the European conference on computer vision (ECCV), 2018, pp. 801–818

  4. [4]

    Unet++: Redesigning skip connections to exploit multiscale features in image segmentation,

    Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, “Unet++: Redesigning skip connections to exploit multiscale features in image segmentation,” IEEE transactions on medical imaging , vol. 39, no. 6, pp. 1856–1867, 2019

  5. [5]

    Lvit: language meets vision transformer in medical image segmentation,

    Z. Li, Y . Li, Q. Li, P. Wang, D. Guo, L. Lu, D. Jin, Y . Zhang, and Q. Hong, “Lvit: language meets vision transformer in medical image segmentation,” IEEE transactions on medical imaging , 2023

  6. [6]

    H2former: An efficient hierarchical hybrid transformer for medical image segmentation,

    A. He, K. Wang, T. Li, C. Du, S. Xia, and H. Fu, “H2former: An efficient hierarchical hybrid transformer for medical image segmentation,” IEEE Transactions on Medical Imaging , vol. 42, no. 9, pp. 2763–2775, 2023

  7. [7]

    Uncertainty-aware hierarchical aggregation network for medical image segmentation,

    T. Zhou, Y . Zhou, G. Li, G. Chen, and J. Shen, “Uncertainty-aware hierarchical aggregation network for medical image segmentation,”IEEE Transactions on Circuits and Systems for Video Technology , 2024

  8. [8]

    Erdunet: An efficient residual double- coding unet for medical image segmentation,

    H. Li, D.-H. Zhai, and Y . Xia, “Erdunet: An efficient residual double- coding unet for medical image segmentation,” IEEE Transactions on Circuits and Systems for Video Technology , 2024

Show all 51 references
  1. [9]

    Hunting sparsity: Density- guided contrastive learning for semi-supervised semantic segmentation,

    X. Wang, B. Zhang, L. Yu, and J. Xiao, “Hunting sparsity: Density- guided contrastive learning for semi-supervised semantic segmentation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 3114–3123

  2. [10]

    Augmentation matters: A simple-yet-effective approach to semi-supervised semantic segmentation,

    Z. Zhao, L. Yang, S. Long, J. Pi, L. Zhou, and J. Wang, “Augmentation matters: A simple-yet-effective approach to semi-supervised semantic segmentation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2023, pp. 11 350–11 359

  3. [11]

    Revisiting weak- JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 10 to-strong consistency in semi-supervised semantic segmentation,

    L. Yang, L. Qi, L. Feng, W. Zhang, and Y . Shi, “Revisiting weak- JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 10 to-strong consistency in semi-supervised semantic segmentation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition,...

  4. [12]

    Corrmatch: Label propagation via correlation matching for semi-supervised semantic segmentation,

    B. Sun, Y . Yang, L. Zhang, M.-M. Cheng, and Q. Hou, “Corrmatch: Label propagation via correlation matching for semi-supervised semantic segmentation,” in Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , 2024, pp. 3097–3107

  5. [13]

    Fixmatch: Simplifying semi- supervised learning with consistency and confidence,

    K. Sohn, D. Berthelot, N. Carlini, Z. Zhang, H. Zhang, C. A. Raffel, E. D. Cubuk, A. Kurakin, and C.-L. Li, “Fixmatch: Simplifying semi- supervised learning with consistency and confidence,” Advances in neural information processing systems , vol. 33, pp. 596–608, 2020

  6. [14]

    Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks,

    D.-H. Lee et al., “Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks,” in Workshop on challenges in representation learning, ICML , vol. 3, no. 2. Atlanta, 2013, p. 896

  7. [15]

    Pseudo-labeling and confirmation bias in deep semi-supervised learn- ing,

    E. Arazo, D. Ortego, P. Albert, N. E. O’Connor, and K. McGuinness, “Pseudo-labeling and confirmation bias in deep semi-supervised learn- ing,” in 2020 International joint conference on neural networks (IJCNN). IEEE, 2020, pp. 1–8

  8. [16]

    Segment anything in medical images,

    J. Ma, Y . He, F. Li, L. Han, C. You, and B. Wang, “Segment anything in medical images,” Nature Communications, vol. 15, no. 1, p. 654, 2024

  9. [17]

    Pranet: Parallel reverse attention network for polyp segmentation,

    D.-P. Fan, G.-P. Ji, T. Zhou, G. Chen, H. Fu, J. Shen, and L. Shao, “Pranet: Parallel reverse attention network for polyp segmentation,” in International conference on medical image computing and computer- assisted intervention. Springer, 2020, pp. 263–273

  10. [18]

    Boundary constraint network with cross layer feature integration for polyp segmen- tation,

    G. Yue, W. Han, B. Jiang, T. Zhou, R. Cong, and T. Wang, “Boundary constraint network with cross layer feature integration for polyp segmen- tation,” IEEE Journal of Biomedical and Health Informatics , vol. 26, no. 8, pp. 4090–4099, 2022

  11. [19]

    Acl-net: semi-supervised polyp segmentation via affinity contrastive learning,

    H. Wu, W. Xie, J. Lin, and X. Guo, “Acl-net: semi-supervised polyp segmentation via affinity contrastive learning,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 37, no. 3, 2023, pp. 2812–2820

  12. [20]

    Polyper: Boundary sensitive polyp segmentation,

    H. Shao, Y . Zhang, and Q. Hou, “Polyper: Boundary sensitive polyp segmentation,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 5, 2024, pp. 4731–4739

  13. [21]

    Pseudo labeling methods for semi-supervised semantic segmentation: A review and future perspec- tives,

    L. Ran, Y . Li, G. Liang, and Y . Zhang, “Pseudo labeling methods for semi-supervised semantic segmentation: A review and future perspec- tives,” IEEE Transactions on Circuits and Systems for Video Technology, 2024

  14. [22]

    Semi- supervised medical image segmentation using adversarial consistency learning and dynamic convolution network,

    T. Lei, D. Zhang, X. Du, X. Wang, Y . Wan, and A. K. Nandi, “Semi- supervised medical image segmentation using adversarial consistency learning and dynamic convolution network,” IEEE transactions on medical imaging, vol. 42, no. 5, pp. 1265–1277, 2022

  15. [23]

    Co-training with high-confidence pseudo labels for semi-supervised medical image segmentation,

    Z. Shen, P. Cao, H. Yang, X. Liu, J. Yang, and O. R. Zaiane, “Co-training with high-confidence pseudo labels for semi-supervised medical image segmentation,” arXiv preprint arXiv:2301.04465 , 2023

  16. [24]

    Pseudo-label guided contrastive learning for semi-supervised medical image segmentation,

    H. Basak and Z. Yin, “Pseudo-label guided contrastive learning for semi-supervised medical image segmentation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023, pp. 19 786–19 797

  17. [25]

    Caussl: Causality- inspired semi-supervised learning for medical image segmentation,

    J. Miao, C. Chen, F. Liu, H. Wei, and P.-A. Heng, “Caussl: Causality- inspired semi-supervised learning for medical image segmentation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 21 426–21 437

  18. [26]

    Adaptive bidirectional displace- ment for semi-supervised medical image segmentation,

    H. Chi, J. Pang, B. Zhang, and W. Liu, “Adaptive bidirectional displace- ment for semi-supervised medical image segmentation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2024, pp. 4070–4080

  19. [27]

    Revisiting consistency regularization for semi-supervised change detection in remote sensing images,

    W. G. C. Bandara and V . M. Patel, “Revisiting consistency regularization for semi-supervised change detection in remote sensing images,” arXiv preprint arXiv:2204.08454, 2022

  20. [28]

    Text2seg: Remote sensing image semantic segmentation via text-guided visual foundation models,

    J. Zhang, Z. Zhou, G. Mai, M. Hu, Z. Guan, S. Li, and L. Mu, “Text2seg: Remote sensing image semantic segmentation via text-guided visual foundation models,” arXiv preprint arXiv:2304.10597 , 2023

  21. [29]

    Dynamically updated semi-supervised change detection network combining cross-supervision and screening algorithms,

    S. Yuan, R. Zhong, C. Yang, Q. Li, and Y . Dong, “Dynamically updated semi-supervised change detection network combining cross-supervision and screening algorithms,” IEEE Transactions on Geoscience and Re- mote Sensing, 2024

  22. [30]

    Self-training with noisy student improves imagenet classification,

    Q. Xie, M.-T. Luong, E. Hovy, and Q. V . Le, “Self-training with noisy student improves imagenet classification,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 10 687–10 698

  23. [31]

    Rethinking pre-training and self-training,

    B. Zoph, G. Ghiasi, T.-Y . Lin, Y . Cui, H. Liu, E. D. Cubuk, and Q. Le, “Rethinking pre-training and self-training,” Advances in neural information processing systems , vol. 33, pp. 3833–3845, 2020

  24. [32]

    Meta pseudo labels,

    H. Pham, Z. Dai, Q. Xie, and Q. V . Le, “Meta pseudo labels,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 11 557–11 568

  25. [33]

    Lassl: Label-guided self-training for semi-supervised learning,

    Z. Zhao, L. Zhou, L. Wang, Y . Shi, and Y . Gao, “Lassl: Label-guided self-training for semi-supervised learning,” in Proceedings of the AAAI conference on artificial intelligence, vol. 36, no. 8, 2022, pp. 9208–9216

  26. [34]

    Semi-supervised semantic segmentation with cross pseudo supervision,

    X. Chen, Y . Yuan, G. Zeng, and J. Wang, “Semi-supervised semantic segmentation with cross pseudo supervision,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 2613–2622

  27. [35]

    Conflict- based cross-view consistency for semi-supervised semantic segmenta- tion,

    Z. Wang, Z. Zhao, X. Xing, D. Xu, X. Kong, and L. Zhou, “Conflict- based cross-view consistency for semi-supervised semantic segmenta- tion,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2023, pp. 19 585–19 595

  28. [36]

    St++: Make self-training work better for semi-supervised semantic segmentation,

    L. Yang, W. Zhuo, L. Qi, Y . Shi, and Y . Gao, “St++: Make self-training work better for semi-supervised semantic segmentation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 4268–4277

  29. [37]

    Towards the uncharted: Density-descending feature perturbation for semi-supervised semantic segmentation,

    X. Wang, H. Bai, L. Yu, Y . Zhao, and J. Xiao, “Towards the uncharted: Density-descending feature perturbation for semi-supervised semantic segmentation,” in Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , 2024, pp. 3303–3312

  30. [38]

    Momentum contrast for unsupervised visual representation learning,

    K. He, H. Fan, Y . Wu, S. Xie, and R. Girshick, “Momentum contrast for unsupervised visual representation learning,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 9729–9738

  31. [39]

    A simple framework for contrastive learning of visual representations,

    T. Chen, S. Kornblith, M. Norouzi, and G. Hinton, “A simple framework for contrastive learning of visual representations,” in International conference on machine learning . PMLR, 2020, pp. 1597–1607

  32. [40]

    Bootstrap your own latent-a new approach to self-supervised learning,

    J.-B. Grill, F. Strub, F. Altch ´e, C. Tallec, P. Richemond, E. Buchatskaya, C. Doersch, B. Avila Pires, Z. Guo, M. Gheshlaghi Azar et al. , “Bootstrap your own latent-a new approach to self-supervised learning,” Advances in neural information processing systems, vol. 33, pp. ...

  33. [41]

    Unsupervised learning of visual features by contrasting cluster assign- ments,

    M. Caron, I. Misra, J. Mairal, P. Goyal, P. Bojanowski, and A. Joulin, “Unsupervised learning of visual features by contrasting cluster assign- ments,” Advances in neural information processing systems , vol. 33, pp. 9912–9924, 2020

  34. [42]

    Barlow twins: Self-supervised learning via redundancy reduction,

    J. Zbontar, L. Jing, I. Misra, Y . LeCun, and S. Deny, “Barlow twins: Self-supervised learning via redundancy reduction,” in International conference on machine learning . PMLR, 2021, pp. 12 310–12 320

  35. [43]

    Supervised contrastive learn- ing,

    P. Khosla, P. Teterwak, C. Wang, A. Sarna, Y . Tian, P. Isola, A. Maschinot, C. Liu, and D. Krishnan, “Supervised contrastive learn- ing,” Advances in neural information processing systems , vol. 33, pp. 18 661–18 673, 2020

  36. [44]

    Pixel contrastive-consistent semi-supervised semantic segmentation,

    Y . Zhong, B. Yuan, H. Wu, Z. Yuan, J. Peng, and Y .-X. Wang, “Pixel contrastive-consistent semi-supervised semantic segmentation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 7273–7282

  37. [45]

    C3-semiseg: Contrastive semi-supervised segmentation via cross-set learning and dynamic class-balancing,

    Y . Zhou, H. Xu, W. Zhang, B. Gao, and P.-A. Heng, “C3-semiseg: Contrastive semi-supervised segmentation via cross-set learning and dynamic class-balancing,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2021, pp. 7036–7045

  38. [46]

    Semi-supervised semantic segmentation using unreliable pseudo- labels,

    Y . Wang, H. Wang, Y . Shen, J. Fei, W. Li, G. Jin, L. Wu, R. Zhao, and X. Le, “Semi-supervised semantic segmentation using unreliable pseudo- labels,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2022, pp. 4248–4257

  39. [47]

    Semi-supervised medical image segmentation via hard positives oriented contrastive learning,

    C. Tang, X. Zeng, L. Zhou, Q. Zhou, P. Wang, X. Wu, H. Ren, J. Zhou, and Y . Wang, “Semi-supervised medical image segmentation via hard positives oriented contrastive learning,” Pattern Recognition, vol. 146, p. 110020, 2024

  40. [48]

    Mean teachers are better role mod- els: Weight-averaged consistency targets improve semi-supervised deep learning results,

    A. Tarvainen and H. Valpola, “Mean teachers are better role mod- els: Weight-averaged consistency targets improve semi-supervised deep learning results,” Advances in neural information processing systems , vol. 30, 2017

  41. [49]

    Semi- supervised semantic segmentation needs strong, varied perturbations,

    G. French, S. Laine, T. Aila, M. Mackiewicz, and G. Finlayson, “Semi- supervised semantic segmentation needs strong, varied perturbations,” in British Machine Vision Conference , no. 31, 2020

  42. [50]

    Freematch: Self-adaptive thresholding for semi-supervised learning,

    Y . Wang, H. Chen, Q. Heng, W. Hou, Y . Fan, Z. Wu, J. Wang, M. Savvides, T. Shinozaki, B. Raj et al. , “Freematch: Self-adaptive thresholding for semi-supervised learning,” in Eleventh International Conference on Learning Representations . OpenReview. net, 2023

  43. [51]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 770–778. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 11 Lingcong Cai is currently stu...

Pith tools

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