Pith. sign in

REVIEW 6 major objections 6 minor 39 references

Image Segmentation with transformers: An Overview, Challenges and Future

T0 review · 6 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Self-attention gives image segmentation the global context that CNNs miss

desk verdict A beginner-oriented transformer segmentation survey whose narrative is correct but whose comparative tables, the only real content, contain misattributions and unsupported numbers. read the letter →

arxiv 2501.09372 v1 pith:QD3OUGHX submitted 2025-01-16 cs.CV

Deepjyoti Chetia , Debasish Dutta , Sanjib Kr Kalita This is my paper · ORCID
classification cs.CV
keywords ImagesegmentationTransformerVisionSelf-attentionSemanticInstancePanopticDeeplearning
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

This survey argues that transformer-based architectures overcome the core limitations of convolutional neural networks for image segmentation: CNNs' fixed and slowly growing receptive fields miss global context, while transformers' self-attention lets every image patch relate to every other patch in one step. The paper traces the shift from FCN and U-Net through ViT, DETR, Swin, Segmenter and Mask2Former, framing it as a move from handcrafted components such as anchors and region proposals to end-to-end prediction. The authors present transformer models as handling objects at different scales and capturing context across the whole image, with remaining issues of data hunger, compute cost and interpretability. The survey is intended as a guide to this transition, so its usefulness stands on the accuracy of its model-and-metric tables.

What carries the argument

The load-bearing object is the scaled dot-product self-attention operation, defined as $\mathrm{Attention}(Q,K,V)=\mathrm{softmax}(QK^T/\sqrt{d_k})V$ with $d_k$ the key dimension. The survey's account of why transformers win at segmentation rests on this mechanism allowing any pair of image patches or pixels to interact in a single layer, together with ViT-style patch tokenization that turns an image into linearly projected embeddings plus positional encodings. Variants such as shifted-window attention in Swin and masked attention in Mask2Former adapt this core mechanism to dense prediction while keeping its global-reasoning property.

What would settle it

On a benchmark with clearly long-range dependencies, compare a transformer segmenter with a CNN segmenter whose effective receptive field is expanded by dilated convolutions; if the CNN matches or beats the transformer on those images, then the claim that self-attention is what makes transformers win at global-context segmentation is falsified.

Watch

Extended reading notes

Core claim

The paper's central claim is that transformers, with scaled dot-product self-attention as their core, address three chief shortcomings of CNN-based segmentation: limited receptive fields that grow only with network depth, difficulty with objects at very different scales, and reliance on manually tuned components like anchor boxes, region proposals and feature pyramids. It argues that the attention formula $\mathrm{Attention}(Q,K,V)=\mathrm{softmax}(QK^T/\sqrt{d_k})V$ models long-range dependency directly, so each patch can draw context from the entire image rather than from a local neighborhood. The paper points to Swin, SETR, Segmenter, DETR and Mask2Former as evidence that this mechanism, combined with hierarchical or query-based decoders, delivers state-of-the-art results in semantic, instance and panoptic segmentation. It also positions the end-to-end, anchor-free pipeline of DETR as the removal of architectural complexity that previously hindered generalization.

Load-bearing premise

The load-bearing premise is that the comparison tables accurately record which model achieves which score on which dataset; if key entries are wrong or unsupported, the survey's usefulness as a guide collapses.

Editorial extensions

If this is right

  • Segmentation models can be built as single end-to-end networks, dropping separate region-proposal and anchor stages that had to be tuned by hand.
  • Segmentation quality on large connected objects should improve when the model's attention spans the full image, making global-context coverage a concrete design target.
  • Hierarchical transformer backbones make it practical to handle small and large objects without relying on multi-scale image pyramids.
  • Data efficiency and computational cost become the next bottlenecks, since the transformer segmenters the paper surveys are trained on large datasets and are compute-hungry.

Reading between the lines

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

  • The paper does not run it, but its logic implies: a CNN augmented with a single lightweight self-attention layer should come close to a full transformer on long-range segmentation, which would isolate attention as the causal factor.
  • The paper notes compute cost as a challenge; an observable consequence is that segmentation quality should fall off in a predictable way as input resolution rises, because self-attention cost grows quadratically with the number of patches.
  • An extension the survey leaves implicit: the same attention mechanism that ties semantic and instance segmentation together in Mask2Former points toward video panoptic segmentation, where long-range temporal dependencies play an analogous role.
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

6 major / 6 minor

Summary. This manuscript is a review of transformer-based image segmentation. It defines semantic, instance, and panoptic segmentation; summarizes datasets and evaluation metrics; lists CNN-era models (FCN, U-Net, DeepLab, SegNet, Mask R-CNN, PSPNet, DenseNet-based models, HRNet) in Table 2; presents six transformer-era models (ViT, SETR, DETR, Swin, Segmenter, Mask2Former) in Table 3; and argues in Section 3.3 that self-attention overcomes key CNN limitations: restricted receptive fields, difficulty with multi-scale objects, and reliance on handcrafted components such as anchors and region proposals. The paper contains no experiments, derivations, or code; its evidentiary content consists of the two comparative tables and a narrative synthesis of results already published elsewhere.

Significance. If the comparison tables were accurate, this would be a serviceable tutorial reference: the exposition of the attention mechanism, the task taxonomy, and the dataset overview are clear, and the bibliography covers the central primary sources (Vaswani et al.; ViT; DETR; Swin; Segmenter; Mask2Former; the main CNN baselines). The paper ships no reproducible artifacts, machine-checked proofs, or falsifiable predictions; its scientific contribution is entirely expository. The high-level thesis that self-attention captures global context and removes handcrafted components is consistent with the consensus literature, but it restates established results rather than adding critical analysis. Because the paper's only evidence-bearing content is Tables 2 and 3, the survey's value stands or falls on their factual reliability, and the errors detailed below are pervasive.

major comments (6)
  1. [§3.2, Table 3 (ViT row)] The ViT row reports 'mIoU ~72% on ADE20K' and cites reference [21]; the cited paper (Dosovitskiy et al., 2020) presents image classification experiments only and reports no semantic segmentation result on ADE20K. The entry is therefore unsupported by its source, and the advantage column's claim of strong global feature representation is not backed by any comparable segmentation result in Table 3.
  2. [§3.2, Table 3 (SETR and Segmenter rows)] The row labeled 'SETR (Segmenter) [33]' conflates two distinct architectures: SETR (Zheng et al., ref. [33]) and Segmenter (Strudel et al., ref. [29]) have different encoder-decoder designs, and the key-feature descriptions are interchanged between the two rows. In addition, the Segmenter row reports 'mIoU ~80% on Pascal VOC', whereas Segmenter's published evaluations are on ADE20K, Cityscapes, and Pascal Context. A comparative table that merges two models and misstates a benchmark cannot function as a reference for either architecture.
  3. [§3.2, Table 3 (DETR row)] The DETR row reports 'mIoU ~44% on the COCO dataset' with reference [22], but the original DETR paper evaluates object detection with COCO AP and panoptic segmentation with PQ; it does not report a COCO mIoU. The row's description 'extended for instance segmentation using bipartite matching' also misstates the paper's panoptic segmentation extension. The claimed advantage that DETR 'integrates detection and segmentation seamlessly' is not supported by the cited results as presented.
  4. [§3.2, Table 3 (Swin Transformer row)] The Swin Transformer row reports 'mIoU ~84.1% on ADE20K'. Published state-of-the-art mIoU values on ADE20K are in the 50s to mid-60s, so this figure is not plausible for that benchmark and appears to reflect a metric or dataset mismatch. This entry, like the others in Table 3, needs to be re-derived from the primary source with the correct dataset, split, and metric.
  5. [§2.5, Table 2 (Mask R-CNN row)] The Mask R-CNN row is cited as '[he2017]', but no such entry appears in the reference list (ref. [13] is the ResNet paper), so the reported 'AP of 37.1%' cannot be verified by the reader. The same row reports 'mIoU ~50%' on COCO, although COCO instance segmentation is standardly evaluated with mask AP rather than mIoU. Table 2 should carry a resolvable citation for every row and a consistent metric per task.
  6. [§3.2, Table 3 (comparability across rows)] Even setting aside the misattributed entries, Table 3 compares models across four different benchmarks with two different metrics (mIoU and AP) and no common backbone or baseline, so no two rows are directly comparable. The caption and the bookending text present these as 'SOTA' results, which implicitly invites a comparative reading that the table's structure cannot support. Either a common benchmark/metric should be used per task, or the paper should explicitly disclaim cross-row comparability.
minor comments (6)
  1. [§3 (heading)] The heading 'Trasformer Network' is a typo for 'Transformer Network'.
  2. [§2.5, Table 2] Citation style is inconsistent: Table 2 uses author-date style ('[he2017]') while the rest of the paper uses numbered references; unify the style and ensure every citation resolves to a reference-list entry.
  3. [§2.3, Table 1] Table 1 lists BRATS with year 2012, while the cited reference [24] is the 2015 Multimodal Brain Tumor Image Segmentation Benchmark paper; align the year with the source.
  4. [§2.1, Figure 1] Figure 1 is referenced in the text but is not present in the manuscript as provided; confirm that the figure renders in the compiled version.
  5. [§3.2] The criteria for selecting the six models in Table 3 and the literature coverage period of the survey are not stated, so the representativeness of the SOTA selection cannot be assessed.
  6. [§4] The 'Challenges and Future Trends' section is a single paragraph that lists four challenges with little technical elaboration and no discussion of concrete proposed solutions; for a paper whose title advertises challenges and future directions, this section should be expanded and tied to the surveyed models.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the survey summarizes external transformer-segmentation results and derives nothing from its own inputs.

full rationale

This is a literature survey with no derivation chain of its own: it contains no fitted parameters, no prediction generated from a fitted input, and no architectural claim that is justified by the authors' prior work. The central assertion that transformers with self-attention capture global dependencies better than CNNs is a restatement of results already established in the cited primary papers, such as ViT [21] and DETR [22], not a new result derived from the survey's own assumptions. The paper contains no equation or definition in which an output is defined in terms of a target result, and it makes no self-citations: none of the references [1]-[35] lists any of the authors. The factual errors in Tables 2 and 3 flagged by the reader, such as labeling SETR as 'SETR (Segmenter)', crediting DETR with a segmentation mIoU on COCO, and citing '[he2017]' without a matching bibliography entry, are correctness and sourcing problems, not circularity: they do not make any claimed result equivalent to its inputs by construction. Because the survey is self-contained as a summary of external published findings and performs no independent derivation, the circularity score is 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

A survey introduces no fitted parameters or invented entities. The load-bearing assumptions are that the cited benchmark numbers are accurate and that the model selection is representative; both are questionable in this preprint.

assumptions (3)
  • domain assumption CNNs have limited receptive fields that grow slowly with depth, so they struggle with long-range dependencies and global context.
    Invoked in Section 3.3 to motivate the transformer advantage; cited from [12] and [21], treated as established fact.
  • domain assumption The benchmark numbers reported in Tables 2 and 3 accurately represent the cited models' performance.
    The survey's comparative claims rely on these numbers, but several entries are unverified or incorrect (e.g., DETR mIoU on COCO, SETR/Segmenter labeling).
  • ad hoc to paper The six models in Table 3 are a representative selection of state-of-the-art transformer segmentation models.
    No inclusion criteria, search strategy, or date cutoff is given in Section 3.2; the selection appears ad hoc.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Image Segmentation with transformers: An Overview, Challenges and Future." pith.science (2026). https://pith.science/paper/QD3OUGHX

@misc{pith2026250109372,
  author       = {Pith},
  title        = {Pith review of: Image Segmentation with transformers: An Overview, Challenges and Future},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QD3OUGHX}},
  note         = {Machine review of arXiv:2501.09372}
}
read the original abstract

Image segmentation, a key task in computer vision, has traditionally relied on convolutional neural networks (CNNs), yet these models struggle with capturing complex spatial dependencies, objects with varying scales, need for manually crafted architecture components and contextual information. This paper explores the shortcomings of CNN-based models and the shift towards transformer architectures -to overcome those limitations. This work reviews state-of-the-art transformer-based segmentation models, addressing segmentation-specific challenges and their solutions. The paper discusses current challenges in transformer-based segmentation and outlines promising future trends, such as lightweight architectures and enhanced data efficiency. This survey serves as a guide for understanding the impact of transformers in advancing segmentation capabilities and overcoming the limitations of traditional models.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 10 canonical work pages

  1. [21]

    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,

    A. Dosovitskiy et al. , “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” 2020, doi: 10.48550/ARXIV.2010.11929

    arXiv doi:10.48550/arxiv.2010.11929 2020

  2. [33]

    Rethinking Semantic Segmentation from a Sequence -to-Sequence Perspective with Transformers

    S. Zheng et al. , “Rethinking Semantic Segmentation from a Sequence -to-Sequence Perspective with Transformers.” arXiv, Jul. 2021. Accessed: Oct. 29, 2024. [Online]. Available: https://arxiv.org/abs/2012.15840

    arXiv 2021

  3. [29]

    Segmenter: Transformer for Semantic Segmentation

    R. Strudel, R. Garcia, I. Laptev, and C. Schmid, “Segmenter: Transformer for Semantic Segmentation.” arXiv, Sep. 2021. Accessed: Nov. 21, 2024. [Online]. Available: https://arxiv.org/abs/2105.05633

    arXiv 2021

  4. [22]

    End -to-End Object Detection with Transformers

    N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov, and S. Zagoruyko, “End -to-End Object Detection with Transformers.” arXiv, May 2020. Accessed: Oct. 29, 2024. [Online]. Available: https://arxiv.org/abs/2005.12872

    arXiv 2020

  5. [13]

    Deep Residual Learning for Image Recognition

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition. ” arXiv, Dec

    work page

  6. [1]

    A Review on the Strategies and Techniques of Image Segmentation,

    A. Bali and S. N. Singh, “A Review on the Strategies and Techniques of Image Segmentation,” in 2015 Fifth International Conference on Advanced Computing & Communication Technologies , Haryana, India: IEEE, Feb. 2015, pp. 113–120. doi: 10.1109/ACCT.2015.63

    work page doi:10.1109/acct.2015.63 2015

  7. [2]

    Techniques and Challenges of Image Segmentation: A Review,

    Y. Yu et al., “Techniques and Challenges of Image Segmentation: A Review,” Electronics, vol. 12, no. 5, p. 1199, Mar. 2023, doi: 10.3390/electronics12051199

    work page doi:10.3390/electronics12051199 2023

  8. [3]

    Fu lly convolutional networks for semantic segmentation,

    J. Long, E. Shelhamer, and T. Darrell, “Fu lly convolutional networks for semantic segmentation,” in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , Boston, MA, USA: IEEE, Jun. 2015, pp. 3431–3440. doi: 10.1109/CVPR.2015.7298965

    arXiv 2015

Show all 39 references
  1. [4]

    U -Net: Convolutional Networks for Biomedical Image Segmentation

    O. Ronneberger, P. Fischer, and T. Brox, “U -Net: Convolutional Networks for Biomedical Image Segmentation.” arXiv, May 2015. Accessed: Nov. 12, 2024. [Online]. Available: https://arxiv.org/abs/1505.04597

    2015 arXiv

  2. [5]

    Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs

    L.-C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille, “Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs.” arXiv, Jun. 2016. Accessed: Nov. 21, 2024. [Online]. Available: https://arxiv.org/abs/1412.7062

    2016 arXiv

  3. [6]

    Pyramid Scene Parsing Network

    H. Zhao, J. Shi, X. Qi, X. Wang, and J. Jia, “Pyramid Scene Parsing Network.” arXiv, Apr. 2017. Accessed: Nov. 16, 2024. [Online]. Available: https://arxiv.org/abs/1612.01105

    2017 arXiv

  4. [7]

    ImageNet: A large -scale hierarchical image database,

    J. Deng, W. Dong, R. Socher, L. -J. Li, Kai Li, and Li Fei -Fei, “ImageNet: A large -scale hierarchical image database,” in 2009 IEEE Conference on Computer Vision and Pattern Recognition , Miami, FL: IEEE, Jun. 2009, pp. 248–255. doi: 10.1109/CVPR.2009.5206848

    2009

  5. [8]

    The Pascal Visual Object Classes (VOC) Challenge,

    M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, “The Pascal Visual Object Classes (VOC) Challenge,” International Journal of Computer Vision , vol. 88, no. 2, pp. 303 –338, Jun. 2010, doi: 10.1007/s11263-009-0275-4

    2010 doi

  6. [9]

    Microsoft COCO: Common Objects in Context,

    T.-Y. Lin et al., “Microsoft COCO: Common Objects in Context,” in Computer Vision – ECCV 2014, vol. 8693, D. Fleet, T. Pajdla, B. Schiele, and T. Tuytelaars, Eds., Cham: Springer International Publishing, 2014, pp. 740–755. doi: 10.1007/978-3-319-10602-1_48

    2014 doi

  7. [10]

    The Cityscapes Dataset for Semantic Urban Scene Understanding

    M. Cordts et al., “The Cityscapes Dataset for Semantic Urban Scene Understanding.” arXiv, Apr

  8. [11]

    The Liver Tu mor Segmentation Benchmark (LiTS),

    P. Bilic et al., “The Liver Tu mor Segmentation Benchmark (LiTS),” Medical Image Analysis , vol. 84, p. 102680, Feb. 2023, doi: 10.1016/j.media.2022.102680

    2023

  9. [12]

    Multi -Scale Context Aggregation by Dilat ed Convolutions

    F. Yu and V. Koltun, “Multi -Scale Context Aggregation by Dilat ed Convolutions.” arXiv, Apr. 2016. Accessed: Nov. 18, 2024. [Online]. Available: https://arxiv.org/abs/1511.07122

    2016 arXiv

  10. [14]

    Dropout: A Simple Way to Prevent Neural Networks from Overfitting,

    N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov , “Dropout: A Simple Way to Prevent Neural Networks from Overfitting,” Journal of Machine Learning Research , vol. 15, no. 56, pp. 1929–1958, 2014

    1929

  11. [15]

    Attention Is All You Need,

    A. Vaswani et al., “Attention Is All You Need,” 2017, doi: 10.48550/ARXIV.1706.03762

    2017 doi arXiv

  12. [16]

    Long Short -Term Memory,

    S. Hochreiter and J. Schmidhuber, “Long Short -Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, Nov. 1997, doi: 10.1162/neco.1997.9.8.1735

    1997 doi

  13. [17]

    Non -local Neural Networks

    X. Wang, R. Girshick, A. Gupta, and K. He, “Non -local Neural Networks.” arXiv, Apr. 2018. Accessed: Nov. 18, 2024. [Online]. Available: https://arxiv.org/abs/1711.07971

    2018 arXiv

  14. [18]

    PSANet: Point-wise Spatial Attention Network for Scene Parsing,

    H. Zhao et al., “PSANet: Point-wise Spatial Attention Network for Scene Parsing,” in Computer Vision – ECCV 2018 , vol. 11213, V. Ferrari, M. Hebert, C. Sminchisescu, and Y. Weiss, Eds., Cham: Springer International Publishing, 2018, pp. 270–286. doi: 10.1007/978-3-030-01240-3_17

    2018 doi

  15. [19]

    Local Relation Networks for Image Recognition

    H. Hu, Z. Zhang, Z. Xie, and S. Lin, “Local Relation Networks for Image Recognition.” arXiv, Apr

  16. [20]

    Exploring Self -attention for Image Recognition

    H. Zhao, J. Jia, and V. Koltun, “Exploring Self -attention for Image Recognition.” arXiv, Apr. 2020. Accessed: Nov. 19, 2024. [Online]. Available: https://arxiv.org/abs/2004.13621

    2020 arXiv

  17. [23]

    Scene Parsing through ADE20K Dataset,

    B. Zhou, H. Zhao, X. Puig, S. Fidler, A. Barriuso, and A. Torralba, “Scene Parsing through ADE20K Dataset,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , Honolulu, HI: IEEE, Jul. 2017, pp. 5122–5130. doi: 10.1109/CVPR.2017.544

    2017 doi

  18. [24]

    The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS),

    B. H. Menze et al., “The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS),” IEEE Transactions on Medical Imaging, vol. 34, no. 10, pp. 1993–2024, Oct. 2015, doi: 10.1109/TMI.2014.2377694

    1993

  19. [25]

    CNN or RNN: Review and Experimental Comparison on Image Classification,

    C. Junliang, “CNN or RNN: Review and Experimental Comparison on Image Classification,” in 2022 IEEE 8th International Conference on Computer and Communications (ICCC), Chengdu, China: IEEE, Dec. 2022, pp. 1939–1944. doi: 10.1109/ICCC56324.2022.10065984

    2022

  20. [26]

    SegNet : A Deep Convolutional Encoder -Decoder Architecture for Image Segmentation

    V. Badrinarayanan, A. Kendall, and R. Cipolla, “SegNet : A Deep Convolutional Encoder -Decoder Architecture for Image Segmentation.” arXiv, Oct. 2016. Accessed: Nov. 21, 2024. [Online]. Available: https://arxiv.org/abs/1511.00561

    2016 arXiv

  21. [27]

    Densely Connected Convolutional Networks

    G. Huang, Z. Liu, L. van der Maaten, and K. Q. Weinberger, “Densely Connected Convolutional Networks.” arXiv, Jan. 2018. Accessed: Nov. 21, 2024. [Online]. Available: https://arxiv.org/abs/1608.06993

    2018 arXiv

  22. [28]

    Deep High-Resolution Representation Learning for Visual Recognition

    J. Wang et al., “Deep High-Resolution Representation Learning for Visual Recognition.” arXiv, Mar

  23. [30]

    Fully Transformer Networks for Semantic Image Segmentation

    S. Wu, T. Wu, F. Lin, S. Tian, and G . Guo, “Fully Transformer Networks for Semantic Image Segmentation.” arXiv, Dec. 2021. Accessed: Nov. 21, 2024. [Online]. Available: https://arxiv.org/abs/2106.04108

    2021 arXiv

  24. [31]

    Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions

    W. Wang et al. , “Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions.” arXiv, Aug. 2021. Accessed: Oct. 29, 2024. [Online]. Available: https://arxiv.org/abs/2102.12122

    2021 arXiv

  25. [32]

    Masked -attention Mask Transformer for Universal Image Segmentation

    B. Cheng, I. Misra, A. G. Schwing, A. Kirillov, and R. Girdhar, “Masked -attention Mask Transformer for Universal Image Segmentation.” arXiv, Jun. 2022. Accessed: Nov. 21, 2024. [Online]. Available: https://arxiv.org/abs/2112.01527

    2022 arXiv

  26. [34]

    Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

    Z. Liu et al. , “Swin Transformer: Hierarchical Vision Transformer using Shifted Windows.” arXiv, Aug. 2021. Accessed: Oct. 29, 2024. [Online]. Available: https://arxiv.org/abs/2103.14030

    2021 arXiv

  27. [35]

    SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers

    E. Xie, W. Wang, Z. Yu, A. Anandkumar, J. M. Alvarez, and P. Luo, “SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers.” arXiv, Oct. 2021. Accessed: Oct. 29, 2024. [Online]. Available: https://arxiv.org/abs/2105.15203

    2021 arXiv

  28. [2015]

    18, 2024

    Accessed: Nov. 18, 2024. [Online]. Available: https://arxiv.org/abs/1512.03385

    2024 arXiv

  29. [2016]

    21, 2024

    Accessed: Nov. 21, 2024. [Online]. Available: https://arxiv.org/abs/1604.01685

    2024 arXiv

  30. [2019]

    19, 2024

    Accessed: Nov. 19, 2024. [Online]. Available: https://arxiv.org/abs/1904.11491

    2024 arXiv

  31. [2020]

    21, 2024

    Accessed: Nov. 21, 2024. [Online]. Available: https://arxiv.org/abs/1908.07919

    2024 arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.