Pith. sign in

REVIEW 4 major objections 6 minor 2 cited by

Feedback-Driven Pseudo-Label Reliability Assessment: Redefining Thresholding for Semi-Supervised Semantic Segmentation

T0 review · 4 major / 6 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 ENCORE replaces static confidence thresholds with class-aware, feedback-driven cutoffs for pseudo-labeling and improves semi-supervised semantic segmentation, particularly when labeled data are scarce.

desk verdict ENCORE is a useful adaptive thresholding scheme for low-data medical segmentation, but the paper doesn't yet prove the feedback signal does the work—send it to review with a request for a random-selection control. read the letter →

arxiv 2505.07691 v1 pith:SGDVIRTA submitted 2025-05-12 cs.CV

Negin Ghamsarian , Sahar Nasirihaghighi , Klaus Schoeffmann , Raphael Sznitman This is my paper · ORCID
classification cs.CV
keywords semi-supervisedlearningsemanticsegmentationpseudo-labelingconfidencethresholdingadaptivethresholdsclass-awarecalibrationmedicalimagefeedback-drivenselection
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 semantic segmentation often trains on pseudo-labels filtered by a fixed confidence threshold, but choosing that threshold needs a large labeled validation set and is especially unreliable when labeled data are scarce. The paper proposes ENCORE, a two-part strategy that sets class-wise thresholds from the model's true-positive confidence on labeled data and then adapts those thresholds during training by testing three candidate thresholds on copies of the student and keeping the one that scores best on the labeled batch. The claim is that this feedback loop keeps informative pseudo-labels while discarding noisy ones, without manual threshold tuning. If correct, it would remove a major hyperparameter from pseudo-supervision frameworks and give the largest gains exactly where labeled data are hardest to obtain.

What carries the argument

The central object is ENCORE, an ensemble of three class-wise threshold controllers coupled to the student model. Its first component, Class-Aware Confidence Calibration (CAC), computes for each class the average confidence of true-positive predictions on the labeled set and uses this vector R to initialize the thresholds, so classes with inherently lower confidence are not over-filtered. Its second component, Adaptive Confidence Thresholding (ACT), creates three assessor networks by copying the current student weights, trains each on the pseudo-labeled batch with a different threshold controller, and selects the controller maximizing Dice on the labeled batch (Eq. 4); controller values are updated only after the same controller is selected n consecutive times (Eq. 5). The assessor networks are the mechanism that carries the argument: they turn threshold selection into an online, metric-driven search rather than a fixed hyperparameter.

What would settle it

Measure, across training iterations, the rank correlation between the Dice score on the labeled batch used in Eq. (4) and the Dice score on a held-out validation set; if this correlation is not positive in the low-data splits, the selected thresholds are not tracking generalization and the reported gains would not be expected to transfer.

Watch

Extended reading notes

Core claim

The central discovery is that pseudo-label filtering does not need a pre-set confidence threshold: a threshold controller can be learned online from the student's own response. ENCORE first computes a class-wise reliability indicator R from the labeled set, the average confidence of true-positive predictions per class, and uses it to initialize three threshold controllers. At each training step it copies the student into three assessor networks, trains each assessor on the pseudo-labeled batch under a different controller, and selects the controller with the highest Dice on the labeled batch; after repeated selections it shifts the whole trio of thresholds upward or downward. The paper reports that this mechanism improves Dice scores over fixed-threshold baselines across five medical segmentation datasets, three network architectures, and several pseudo-supervision frameworks, with the largest relative gains in the lowest labeled-data fractions.

Load-bearing premise

The feedback loop assumes that the Dice score computed on the small labeled mini-batch is a faithful proxy for how well the model will generalize, and in the lowest-data settings that proxy is based on very few labeled samples.

Editorial extensions

If this is right

  • Static confidence thresholds and manual threshold tuning become unnecessary in pseudo-supervision pipelines; thresholds are initialized per class and adjusted during training.
  • In data-scarce settings the gains are largest: for example, on Cataract-1K with 1/26 labeled data, UniMatch plus ENCORE reaches 83.62 Dice versus 78.07 for UniMatch alone, and with one labeled ACDC volume UniMatch plus ENCORE beats supervised training with ten labeled volumes.
  • The improvement transfers across frameworks (UniMatch, Switch, AD-MT) and architectures (DeepLabV3+, UNet, VNet), so the benefit comes from the thresholding strategy rather than from a particular teacher-student design.
  • Adaptive thresholding also narrows fold-to-fold variance on Cataract-1K compared with static-threshold baselines.

Reading between the lines

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

  • Editorial: The feedback loop is a form of online hyperparameter search over three threshold levels; testing more controllers or a continuous update rule might yield further gains, but the paper only evaluates three.
  • Editorial: In the most extreme low-data splits the labeled batch is very small, so the Dice signal in Eq. (4) could overfit; a stress test would be to swap the labeled mini-batch between iterations and check whether selected thresholds remain stable.
  • Editorial: Because the evaluator is a generic performance metric, the same feedback idea could be applied to other pseudo-label filters (entropy, teacher disagreement) or to other dense prediction tasks, though the paper does not test these.
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

4 major / 6 minor

Summary. The manuscript proposes ENCORE, a feedback-driven mechanism for selecting confidence thresholds in semi-supervised semantic segmentation. ENCORE first computes class-wise reliability scores from the labeled set (CAC) and then uses three assessor networks, each applying a different threshold controller, to dynamically select the threshold that maximizes Dice on the labeled batch (ACT). The authors integrate ENCORE into UniMatch, Switch, and AD-MT, and report consistent improvements across five medical imaging datasets, especially in low-data regimes. The central claim is that ENCORE substantially improves existing pseudo-supervision methods in data-scarce conditions.

Significance. If the empirical results hold, the paper addresses a practical and understudied problem: how to set pseudo-label confidence thresholds without large validation sets. The method is simple to describe, integrates with several existing frameworks, and the reported gains in low-data regimes (e.g., Cataract-1K 1/26 and ACDC 1/140) are large. The ablation study isolating CAC and ACT is useful and supports the design. The main weakness is that the threshold-selection signal, Eq. (4), is an in-sample Dice score on a very small labeled batch, and the paper provides no analysis of whether this signal reliably predicts generalization; this is especially concerning in the extreme low-data settings where the method claims its largest gains. No code or checkpoints are provided, and at least one reported metric (Table 5, Rel. Avg.) is internally inconsistent. The core idea is non-circular, however, because threshold selection is evaluated on labeled data rather than on the unlabeled pseudo-labels being filtered.

major comments (4)
  1. [§3.2, Eq. (4)] The threshold selection signal is the Dice score on the labeled mini-batch after a single assessor training step, with assessors copied from a student that has already been trained on that same labeled batch. This is an in-sample, potentially high-variance criterion, yet the paper does not analyze its reliability. In the ACDC 1/140 setting the labeled set is a single volume, and in Cataract-1K 1/26 it is 61 images, precisely the regimes where the paper claims the largest gains. Please provide evidence that the selected threshold controller predicts held-out performance: for example, the distribution of selected controllers over training, the correlation between Eq. (4)'s Dice and validation Dice, or an ablation that selects among the three controllers randomly rather than by Eq. (4). Without such evidence, the reported gains could be an artifact of overfitting the threshold to the labeled batch.
  2. [Tables 4 and 5 (ACDC, LA)] The extreme low-data results (ACDC 1/70 and 1/140; LA 1/26 and 1/40) are reported as single point estimates with no standard deviations, number of runs, or seeds. These are precisely the settings where the proxy used in Eq. (4) is least reliable and where the claimed improvements are largest. The paper states that the method "significantly improves" performance, but without variance estimates the statistical support for that claim in these regimes is not established. Please report multiple seeds or otherwise quantify the variability of these numbers.
  3. [Table 5, Rel. Avg. for AD-MT + ENCORE2] The reported relative average Dice of 8.36 for AD-MT + ENCORE2 in the LA dataset is inconsistent with the per-split values in the same row: averaging the per-split relative improvements over the supervised baseline would yield a value an order of magnitude larger. This suggests either a typo or an inconsistent computation of the relative average metric. Please correct the value and ensure the same formula is used consistently across Tables 4 and 5.
  4. [§3.2, Eq. (5) and Section 4 implementation details] The update rule depends on the parameters n, α1, and α2, which are fixed to n=5 and (α1,α2)=(0.98,1.02) across all datasets and settings. The text claims the method eliminates manual threshold tuning, but these meta-parameters still require manual choice, and no sensitivity analysis is reported. Please report experiments varying n and α1/α2, or at least provide a justification for the fixed values.
minor comments (6)
  1. [Section 4, Cataract-1K description] The text states that the labeled splits correspond to [1/2, 1/4, 1/8, 1/16], but Table 1 reports a 1/26 split with 61 images; please clarify how the number of training cases relates to the labeled fraction and reconcile the discrepancy.
  2. [Eq. (2) and Algorithm 1] Eq. (2) computes the class-wise reliability indicator using "a network trained solely on labeled images," and Algorithm 1 begins with "Train an assessor network on labeled images" lines 1-2, but the relationship between this initial assessor and the three assessor networks used in ACT is not clarified; please define the initialization and whether the same weights are reused.
  3. [Eq. (4)] The notation Dice(·, ·, τ_k) in Eq. (4) is not formally defined; please specify the metric arguments (prediction, ground truth, threshold) and how the threshold enters the Dice computation, since Dice is normally threshold-independent.
  4. [Section 4, ACDC description] The sentence "while excluding it from our configurations" is confusing in the ACDC dataset paragraph; please rephrase to clarify that the paper follows the settings of [52] or describe the actual split procedure.
  5. [Section 5, Figure 5] The kernel density estimate plots are difficult to interpret without axis labels and units; the qualitative claim that ENCORE "bridges the distribution gap" should be supported by a quantitative metric, such as a distributional distance between labeled and unlabeled feature distributions.
  6. [General reproducibility] The paper states that code and training splits will be released upon acceptance, but no code, checkpoints, or detailed per-fold splits are provided in the supplementary; please include the implementation details needed to reproduce the four-fold results, including any random-seed handling.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; ENCORE is an empirical feedback mechanism evaluated on held-out benchmarks.

full rationale

The paper's central claim is that the ENCORE thresholding mechanism improves pseudo-supervised segmentation when added to existing frameworks. Walking the derivation chain, the CAC component (Eq. 2) computes per-class reliability from the labeled set by averaging the confidence of true-positive predictions. This is an empirical calibration statistic, not a quantity whose definition already contains the reported test Dice. The ACT component (Eq. 4) selects one of three threshold controllers by training assessor copies on pseudo-labeled unlabeled batches and measuring Dice on the labeled batch; Eq. 5 then shifts the controller set based on repeated selections. The selected threshold is not fitted to test performance and no reported test result is used as an input to the selection procedure. The labeled Dice used in Eq. 4 is a separate proximal signal from the unlabeled pseudo-label filtering being evaluated, so the selection is not circular in the sense of predicting a quantity that was used to fit it. The paper does not import any load-bearing uniqueness theorem or prior derivation from the authors' own work; self-citations are limited to the Cataract-1K dataset [14] and standard baselines, none of which justify the method's core mechanism. The in-sample nature of the selection signal and the small labeled batches in low-data regimes raise legitimate empirical-validity concerns, but those are correctness or overfitting risks, not circularity: the paper does not define its claimed improvement in terms of the labeled-batch Dice that drives threshold selection. Therefore no specific reduction from Eq. (X) to its own inputs can be exhibited, and no circular step is identified.

Assumptions & free parameters 3 free parameters · 4 assumptions · 2 invented entities

The method introduces a few free hyperparameters (alpha_1, alpha_2, n) that are fixed globally without sensitivity analysis, plus several domain assumptions about confidence transfer, the reliability of Dice as a proxy on small batches, and the assessor-student relationship. These are reasonable engineering assumptions but are not derived from first principles.

free parameters (3)
  • alpha_1, alpha_2 = 0.98, 1.02
    The threshold adaptors in Eqs. (3) and (5) are fixed to (0.98, 1.02) across all settings. They define the initial search space and the rate of threshold adaptation, yet no ablation or sensitivity analysis is provided.
  • n = 5
    The number of consecutive selections of a threshold controller required before an update in Eq. (5). Set to five without ablation or sensitivity analysis.
  • Number of assessor controllers = 3
    The choice of exactly three threshold controllers, initialized as (alpha_1*R, R, alpha_2*R), is an ad hoc design choice. A different number or spacing of controllers could change results.
assumptions (4)
  • domain assumption Class-wise average true-positive confidence on the labeled set (Eq. 2) is an informative prior for the optimal class-wise threshold on the unlabeled set.
    The entire CAC module relies on the model's confidence statistics on the small labeled set transferring to the unlabeled set. This is plausible but unverified in the paper.
  • domain assumption Dice on the labeled mini-batch is a stable and unbiased selector of the optimal threshold controller (Eq. 4).
    The ACT mechanism selects the threshold controller that maximizes Dice on labeled batches. With very small labeled batches, this estimator is high-variance, yet the paper does not analyze potential instabilities or overfitting.
  • domain assumption Assessor networks, obtained by copying student weights and training on pseudo-labeled batches, provide a faithful estimate of the student's downstream performance.
    The feedback loop assumes that the assessor's performance on the labeled batch is a relevant surrogate for the student's test generalization. The paper does not demonstrate that this proxy is faithful.
  • ad hoc to paper A single ensemble of three fixed threshold controllers is sufficient to cover the space of useful thresholds over the whole training trajectory.
    The choice of exactly three controllers at (0.98, 1.02) of R is a design choice not derived from any principle. The update rule in Eq. (5) only shifts all three controllers multiplicatively, which may limit the achievable threshold space.
invented entities (2)
  • Assessor networks (three copies of the student model)
    purpose: They evaluate candidate thresholds during training and produce the Dice feedback used for threshold selection in ACT (Section 3.2, Eq. 4).
    This is an algorithmic component within the method, not a physical or external entity. Its validity only shows up in the final student performance.
  • Class-wise reliability indicator R
    purpose: Serves as a class-wise initialization and reference for the threshold controllers (Eqs. 2 and 3).
    It is computed entirely from the labeled set and is internal to the method, so there is no independent evidence outside the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Feedback-Driven Pseudo-Label Reliability Assessment: Redefining Thresholding for Semi-Supervised Semantic Segmentation." pith.science (2026). https://pith.science/paper/SGDVIRTA

@misc{pith2026250507691,
  author       = {Pith},
  title        = {Pith review of: Feedback-Driven Pseudo-Label Reliability Assessment: Redefining Thresholding for Semi-Supervised Semantic Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SGDVIRTA}},
  note         = {Machine review of arXiv:2505.07691}
}
read the original abstract

Semi-supervised learning leverages unlabeled data to enhance model performance, addressing the limitations of fully supervised approaches. Among its strategies, pseudo-supervision has proven highly effective, typically relying on one or multiple teacher networks to refine pseudo-labels before training a student network. A common practice in pseudo-supervision is filtering pseudo-labels based on pre-defined confidence thresholds or entropy. However, selecting optimal thresholds requires large labeled datasets, which are often scarce in real-world semi-supervised scenarios. To overcome this challenge, we propose Ensemble-of-Confidence Reinforcement (ENCORE), a dynamic feedback-driven thresholding strategy for pseudo-label selection. Instead of relying on static confidence thresholds, ENCORE estimates class-wise true-positive confidence within the unlabeled dataset and continuously adjusts thresholds based on the model's response to different levels of pseudo-label filtering. This feedback-driven mechanism ensures the retention of informative pseudo-labels while filtering unreliable ones, enhancing model training without manual threshold tuning. Our method seamlessly integrates into existing pseudo-supervision frameworks and significantly improves segmentation performance, particularly in data-scarce conditions. Extensive experiments demonstrate that integrating ENCORE with existing pseudo-supervision frameworks enhances performance across multiple datasets and network architectures, validating its effectiveness in semi-supervised learning.

Figures

Figures reproduced from arXiv: 2505.07691 by the authors.

Figure 1
Figure 1. Comparison of Dice scores across different confidence [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of ENCORE: A feedback-driven pseudo-label refinement framework integrating Class-Aware Confidence Calibration [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Comparison between CAC-based confidence thresholds [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Qualitative comparisons of state-of-the-art methods with and without ENCORE on the Cataract-1K, LA, and ACDC datasets. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Kernel density estimates from various methods trained [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. CW-BASS v2: Saturation-Aware Pseudo-Label Selection for Semi-Supervised Segmentation under Foundation-Model Teachers

    cs.CV 2026-08 conditional novelty 6.0 of 10

    A saturation gate measures the reliability of a teacher's confident pseudo-labels, Pr(correct | confidence >= 0.95), and picks strict filtering when that reliability is high and an adaptive floor when it drops, making...

  2. PixCon: Clean-Positive Contrastive Learning for Foundation-Model Semi-Supervised Segmentation

    cs.CV 2026-07 conditional novelty 5.5 of 10

    A contamination-free (ρ_F=0) labeled-and-correct pixel bank over UniMatch V2 yields a consistent ~+0.2 mIoU per-seed gain on Pascal 1/8 and ties elsewhere, with the gain attributed to correctness sharpening rather tha...

Reference graph

Works this paper leans on

57 extracted references · 44 canonical work pages · cited by 2 Pith papers

  1. [1]

    2017 robotic instrument segmentation challenge

    Max Allan, Alex Shvets, Thomas Kurmann, Zichen Zhang, Rahul Duggal, Yun-Hsuan Su, Nicola Rieke, Iro Laina, Niveditha Kalavakonda, Sebastian Bodenstedt, et al. 2017 robotic instrument segmentation challenge. arXiv preprint arXiv:1902.06426, 2019. 5

    arXiv 2017

  2. [2]

    Bidirectional copy-paste for semi-supervised medical image segmentation

    Yunhao Bai, Duowen Chen, Qingli Li, Wei Shen, and Yan Wang. Bidirectional copy-paste for semi-supervised medical image segmentation. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition , pages 11514–11524, 2023. 6

    work page 2023

  3. [3]

    Deep learning techniques for automatic mri cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE transactions on medical imaging, 37 (11):2514–2525, 2018

    Olivier Bernard, Alain Lalande, Clement Zotti, Freder- ick Cervenansky, Xin Yang, Pheng-Ann Heng, Irem Cetin, Karim Lekadir, Oscar Camara, Miguel Angel Gonzalez Ballester, et al. Deep learning techniques for automatic mri cardiac multi-structures segmentation and diagnosis: is the problem solved? IEEE transactions on medical imaging, 37 (11):2514–2525, 2018. 5

    work page 2018

  4. [4]

    Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolu- tion, and fully connected crfs

    Liang-Chieh Chen, George Papandreou, Iasonas Kokkinos, Kevin Murphy, and Alan L Yuille. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolu- tion, and fully connected crfs. IEEE transactions on pattern analysis and machine intelligence, 40(4):834–848, 2017. 1

    2017

  5. [5]

    Encoder-decoder with atrous separable convolution for semantic image segmentation

    Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam. Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proceedings of the European conference on computer vision (ECCV), pages 801–818, 2018. 1, 6

    work page 2018

  6. [6]

    Semi-supervised domain adaptation based on dual-level domain mixing for semantic segmentation

    Shuaijun Chen, Xu Jia, Jianzhong He, Yongjie Shi, and Jianzhuang Liu. Semi-supervised domain adaptation based on dual-level domain mixing for semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 11018–11027, 2021. 1

    work page 2021

  7. [7]

    Semi-supervised semantic segmentation with cross pseudo supervision

    Xiaokang Chen, Yuhui Yuan, Gang Zeng, and Jingdong Wang. Semi-supervised semantic segmentation with cross pseudo supervision. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 2613–2622, 2021. 2, 3, 6

    work page 2021

  8. [8]

    Mask-based Data Augmentation for Semi-supervised Semantic Segmentation

    Ying Chen, Xu Ouyang, Kaiyue Zhu, and Gady Agam. Mask-based data augmentation for semi-supervised seman- tic segmentation. arXiv preprint arXiv:2101.10156 , 2021. 2

    work page Pith review arXiv 2021

Show all 57 references
  1. [9]

    Interactive network perturbation between teacher and stu- dents for semi-supervised semantic segmentation

    Hyuna Cho, Injun Choi, Suha Kwak, and Won Hwa Kim. Interactive network perturbation between teacher and stu- dents for semi-supervised semantic segmentation. In Pro- ceedings of the IEEE/CVF Winter Conference on Applica- tions of Computer Vision, pages 626–635, 2024. 1, 4, 6

    2024

  2. [10]

    Imagenet: A large-scale hierarchical image database

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee, 2009. 6

    2009

  3. [11]

    Dmt: Dynamic mutual training for semi-supervised learning

    Zhengyang Feng, Qianyu Zhou, Qiqi Gu, Xin Tan, Guan- gliang Cheng, Xuequan Lu, Jianping Shi, and Lizhuang Ma. Dmt: Dynamic mutual training for semi-supervised learning. Pattern Recognition, page 108777, 2022. 2, 3

    2022

  4. [12]

    Semi-supervised semantic segmen- tation needs strong, varied perturbations

    Geoff French, Samuli Laine, Timo Aila, Michal Mackiewicz, and Graham Finlayson. Semi-supervised semantic segmen- tation needs strong, varied perturbations. arXiv preprint arXiv:1906.01916, 2019

    1906 arXiv

  5. [13]

    Dsp: Dual soft-paste for unsupervised domain adaptive semantic segmentation

    Li Gao, Jing Zhang, Lefei Zhang, and Dacheng Tao. Dsp: Dual soft-paste for unsupervised domain adaptive semantic segmentation. In Proceedings of the 29th ACM International Conference on Multimedia, pages 2825–2833, 2021. 2

    2021

  6. [14]

    Cataract-1k: Cataract surgery dataset for scene segmenta- tion, phase recognition, and irregularity detection

    Negin Ghamsarian, Yosuf El-Shabrawi, Sahar Nasiri- haghighi, Doris Putzgruber-Adamitsch, Martin Zinkernagel, Sebastian Wolf, Klaus Schoeffmann, and Raphael Sznitman. Cataract-1k: Cataract surgery dataset for scene segmenta- tion, phase recognition, and irregularity detection. ...

    2023 arXiv

  7. [15]

    Simple copy-paste is a strong data augmentation method for instance segmentation

    Golnaz Ghiasi, Yin Cui, Aravind Srinivas, Rui Qian, Tsung- Yi Lin, Ekin D Cubuk, Quoc V Le, and Barret Zoph. Simple copy-paste is a strong data augmentation method for instance segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition ,...

    2021

  8. [16]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 6

    2016

  9. [17]

    Mask r-cnn

    Kaiming He, Georgia Gkioxari, Piotr Doll ´ar, and Ross Gir- shick. Mask r-cnn. In Proceedings of the IEEE international conference on computer vision, pages 2961–2969, 2017. 1

    2017

  10. [18]

    Semi-supervised semantic segmentation via adaptive equalization learning

    Hanzhe Hu, Fangyun Wei, Han Hu, Qiwei Ye, Jinshi Cui, and Liwei Wang. Semi-supervised semantic segmentation via adaptive equalization learning. Advances in Neural In- formation Processing Systems, 34:22106–22118, 2021. 2

    2021

  11. [19]

    Universal semi-supervised semantic segmenta- tion

    Tarun Kalluri, Girish Varma, Manmohan Chandraker, and CV Jawahar. Universal semi-supervised semantic segmenta- tion. In Proceedings of the IEEE/CVF International Confer- ence on Computer Vision, pages 5259–5270, 2019. 1

    2019

  12. [20]

    Guided collaborative training for pixel-wise semi-supervised learning

    Zhanghan Ke, Di Qiu, Kaican Li, Qiong Yan, and Ryn- son WH Lau. Guided collaborative training for pixel-wise semi-supervised learning. In European conference on com- puter vision, pages 429–445. Springer, 2020. 2

    2020

  13. [21]

    Semi-supervised semantic seg- mentation with directional context-aware consistency

    Xin Lai, Zhuotao Tian, Li Jiang, Shu Liu, Hengshuang Zhao, Liwei Wang, and Jiaya Jia. Semi-supervised semantic seg- mentation with directional context-aware consistency. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 1205–1214, 2021. 2

    2021

  14. [22]

    Temporal ensembling for semi- supervised learning

    Samuli Laine and Timo Aila. Temporal ensembling for semi- supervised learning. CoRR, abs/1610.02242, 2016. 2

    2016 arXiv

  15. [23]

    Pseudo-label: The simple and effi- cient semi-supervised learning method for deep neural net- works

    Dong-Hyun Lee et al. Pseudo-label: The simple and effi- cient semi-supervised learning method for deep neural net- works. In Workshop on challenges in representation learn- ing, ICML, page 896, 2013. 1

    2013

  16. [24]

    Semantic segmentation with generative models: Semi-supervised learning and strong out-of-domain generalization

    Daiqing Li, Junlin Yang, Karsten Kreis, Antonio Torralba, and Sanja Fidler. Semantic segmentation with generative models: Semi-supervised learning and strong out-of-domain generalization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pa...

    2021

  17. [25]

    Bidirectional learning for domain adaptation of semantic segmentation

    Yunsheng Li, Lu Yuan, and Nuno Vasconcelos. Bidirectional learning for domain adaptation of semantic segmentation. In 2019 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 6929–6938, 2019. 1, 2

    2019

  18. [26]

    Microsoft coco: Common objects in context

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Doll´ar, and C Lawrence Zitnick. Microsoft coco: Common objects in context. In European conference on computer vision , pages 740–755. Springer, 2014. 1

    2014

  19. [27]

    Ms- net: Multi-site network for improving prostate segmentation with heterogeneous mri data

    Quande Liu, Qi Dou, Lequan Yu, and Pheng Ann Heng. Ms- net: Multi-site network for improving prostate segmentation with heterogeneous mri data. IEEE Transactions on Medical Imaging, 2020. 5

    2020

  20. [28]

    Perturbed and strict mean teachers for semi-supervised semantic segmenta- tion

    Yuyuan Liu, Yu Tian, Yuanhong Chen, Fengbei Liu, Vasileios Belagiannis, and Gustavo Carneiro. Perturbed and strict mean teachers for semi-supervised semantic segmenta- tion. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 4258–4267,

  21. [29]

    Fully convolutional networks for semantic segmentation

    Jonathan Long, Evan Shelhamer, and Trevor Darrell. Fully convolutional networks for semantic segmentation. In Pro- ceedings of the IEEE conference on computer vision and pat- tern recognition, pages 3431–3440, 2015. 1

    2015

  22. [30]

    Switching temporary teachers for semi-supervised semantic segmentation

    Jaemin Na, Jung-Woo Ha, Hyung Jin Chang, Dongyoon Han, and Wonjun Hwang. Switching temporary teachers for semi-supervised semantic segmentation. Advances in Neural Information Processing Systems, 36, 2024. 1, 2, 4, 6, 7

    2024

  23. [31]

    Classmix: Segmentation-based data aug- mentation for semi-supervised learning

    Viktor Olsson, Wilhelm Tranheden, Juliano Pinto, and Lennart Svensson. Classmix: Segmentation-based data aug- mentation for semi-supervised learning. In Proceedings of the IEEE/CVF Winter Conference on Applications of Com- puter Vision (WACV), pages 1369–1378, 2021. 2

    2021

  24. [32]

    Semi- supervised semantic segmentation with cross-consistency training

    Yassine Ouali, C ´eline Hudelot, and Myriam Tami. Semi- supervised semantic segmentation with cross-consistency training. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 12674– 12684, 2020. 2

    2020

  25. [33]

    Perone, Pedro Ballester, Rodrigo C

    Christian S. Perone, Pedro Ballester, Rodrigo C. Barros, and Julien Cohen-Adad. Unsupervised domain adaptation for medical imaging segmentation with self-ensembling. Neu- roImage, 194:1–11, 2019. 2

    2019

  26. [34]

    Regularization with stochastic transformations and perturba- tions for deep semi-supervised learning

    Mehdi Sajjadi, Mehran Javanmardi, and Tolga Tasdizen. Regularization with stochastic transformations and perturba- tions for deep semi-supervised learning. Advances in neural information processing systems, 29, 2016. 2

    2016

  27. [35]

    Fixmatch: Simplifying semi-supervised learning with consistency and confidence

    Kihyuk Sohn, David Berthelot, Nicholas Carlini, Zizhao Zhang, Han Zhang, Colin A Raffel, Ekin Dogus Cubuk, Alexey Kurakin, and Chun-Liang Li. Fixmatch: Simplifying semi-supervised learning with consistency and confidence. Advances in neural information processing systems, 33:5...

    2020

  28. [36]

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

    Boyuan Sun, Yuqi Yang, Le Zhang, Ming-Ming Cheng, and Qibin Hou. Corrmatch: Label propagation via correlation matching for semi-supervised semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition (CVPR) , pages 3097–3107,

  29. [37]

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

    Antti Tarvainen and Harri Valpola. Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results. Advances in neural information processing systems, 30, 2017. 1, 2, 4, 6

    2017

  30. [38]

    Dacs: Domain adaptation via cross- domain mixed sampling

    Wilhelm Tranheden, Viktor Olsson, Juliano Pinto, and Lennart Svensson. Dacs: Domain adaptation via cross- domain mixed sampling. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pages 1379–1389, 2021. 2

    2021

  31. [39]

    Unsupervised semantic seg- mentation by contrasting object mask proposals

    Wouter Van Gansbeke, Simon Vandenhende, Stamatios Georgoulis, and Luc Van Gool. Unsupervised semantic seg- mentation by contrasting object mask proposals. InProceed- ings of the IEEE/CVF International Conference on Com- puter Vision (ICCV), pages 10052–10062, 2021. 2

    2021

  32. [40]

    Su- dre, Mark S

    Thomas Varsavsky, Mauricio Orbes-Arteaga, Carole H. Su- dre, Mark S. Graham, Parashkev Nachev, and M. Jorge Car- doso. Test-time unsupervised domain adaptation. In Medi- cal Image Computing and Computer Assisted Intervention – MICCAI 2020, pages 428–436, Cham, 2020. Springer I...

    2020

  33. [41]

    Exploring cross-image pixel contrast for semantic segmentation

    Wenguan Wang, Tianfei Zhou, Fisher Yu, Jifeng Dai, En- der Konukoglu, and Luc Van Gool. Exploring cross-image pixel contrast for semantic segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 7303–7313, 2021. 2

    2021

  34. [42]

    Semi-supervised semantic segmentation using unreliable pseudo-labels

    Yuchao Wang, Haochen Wang, Yujun Shen, Jingjing Fei, Wei Li, Guoqiang Jin, Liwei Wu, Rui Zhao, and Xinyi Le. Semi-supervised semantic segmentation using unreliable pseudo-labels. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 4248–...

    2022

  35. [43]

    Mcf: Mutual correction framework for semi- supervised medical image segmentation

    Yongchao Wang, Bin Xiao, Xiuli Bi, Weisheng Li, and Xinbo Gao. Mcf: Mutual correction framework for semi- supervised medical image segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 15651–15660, 2023. 4, 6

    2023

  36. [44]

    Crest: A class-rebalancing self-training frame- work for imbalanced semi-supervised learning

    Chen Wei, Kihyuk Sohn, Clayton Mellina, Alan Yuille, and Fan Yang. Crest: A class-rebalancing self-training frame- work for imbalanced semi-supervised learning. In Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10857–10866, 2021. 3

    2021

  37. [45]

    Unsupervised data augmentation for consistency training

    Qizhe Xie, Zihang Dai, Eduard Hovy, Thang Luong, and Quoc Le. Unsupervised data augmentation for consistency training. Advances in Neural Information Processing Sys- tems, 33:6256–6268, 2020. 2

    2020

  38. [46]

    A global benchmark of algo- rithms for segmenting the left atrium from late gadolinium- enhanced cardiac magnetic resonance imaging

    Zhaohan Xiong, Qing Xia, Zhiqiang Hu, Ning Huang, Cheng Bian, Yefeng Zheng, Sulaiman Vesal, Nishant Ravikumar, Andreas Maier, Xin Yang, et al. A global benchmark of algo- rithms for segmenting the left atrium from late gadolinium- enhanced cardiac magnetic resonance imaging. M...

    2021

  39. [47]

    St++: Make self-training work better for semi-supervised se- mantic segmentation

    Lihe Yang, Wei Zhuo, Lei Qi, Yinghuan Shi, and Yang Gao. St++: Make self-training work better for semi-supervised se- mantic segmentation. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 4268–4277, 2022. 1, 2, 3, 6

    2022

  40. [48]

    Revisiting weak-to-strong consistency in semi-supervised semantic segmentation

    Lihe Yang, Lei Qi, Litong Feng, Wayne Zhang, and Yinghuan Shi. Revisiting weak-to-strong consistency in semi-supervised semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7236–7246, 2023. 2, 4, 6, 7

    2023

  41. [49]

    Cutmix: Regu- larization strategy to train strong classifiers with localizable features

    Sangdoo Yun, Dongyoon Han, Seong Joon Oh, Sanghyuk Chun, Junsuk Choe, and Youngjoon Yoo. Cutmix: Regu- larization strategy to train strong classifiers with localizable features. In Proceedings of the IEEE/CVF international con- ference on computer vision, pages 6023–6032, 2019. 2, 3

    2019

  42. [50]

    Reciprocal learn- ing for semi-supervised segmentation

    Xiangyun Zeng, Rian Huang, Yuming Zhong, Dong Sun, Chu Han, Di Lin, Dong Ni, and Yi Wang. Reciprocal learn- ing for semi-supervised segmentation. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conference, Strasbourg, France, Septe...

    2021

  43. [51]

    Pyramid scene parsing network

    Hengshuang Zhao, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, and Jiaya Jia. Pyramid scene parsing network. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 2881–2890, 2017. 1

    2017

  44. [52]

    Alternate diverse teaching for semi-supervised medical image segmentation

    Zhen Zhao, Zicheng Wang, Longyue Wang, Dian Yu, Yix- uan Yuan, and Luping Zhou. Alternate diverse teaching for semi-supervised medical image segmentation. In European Conference on Computer Vision , pages 227–243. Springer,

  45. [53]

    Pixel contrastive-consistent semi-supervised semantic segmentation

    Yuanyi Zhong, Bodi Yuan, Hong Wu, Zhiqiang Yuan, Jian Peng, and Yu-Xiong Wang. Pixel contrastive-consistent semi-supervised semantic segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 7273–7282, 2021. 2

    2021

  46. [54]

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

    Yanning Zhou, Hang Xu, Wei Zhang, Bin Gao, and Pheng- Ann 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, pages 7036–7045, 2021. 1

    2021

  47. [55]

    Unet++: Redesigning skip connections to exploit multiscale features in image segmen- tation

    Zongwei Zhou, Md Mahfuzur Rahman Siddiquee, Nima Tajbakhsh, and Jianming Liang. Unet++: Redesigning skip connections to exploit multiscale features in image segmen- tation. IEEE transactions on medical imaging, 39(6):1856– 1867, 2019. 1

    2019

  48. [56]

    Yang Zou, Zhiding Yu, B. V . K. Vijaya Kumar, and Jinsong Wang. Unsupervised domain adaptation for semantic seg- mentation via class-balanced self-training. In Computer Vi- sion – ECCV 2018 , pages 297–313, Cham, 2018. Springer International Publishing. 2, 3

    2018

  49. [57]

    Pseudoseg: Designing pseudo labels for semantic segmentation

    Yuliang Zou, Zizhao Zhang, Han Zhang, Chun-Liang Li, Xiao Bian, Jia-Bin Huang, and Tomas Pfister. Pseudoseg: Designing pseudo labels for semantic segmentation. arXiv preprint arXiv:2010.09713, 2020. 2, 3

    2010 arXiv

Pith tools

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