Pith. sign in

REVIEW 4 major objections 7 minor 27 references

Topology-Constrained Learning for Efficient Laparoscopic Liver Landmark Detection

T0 review · 4 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Explicit topological supervision lifts laparoscopic liver landmark detection to state-of-the-art accuracy while cutting inference time to about a third of the closest rival's.

desk verdict A plausible integration of known topology-preserving losses, but the unexplained 5.4-point DSC gap between the main table and the ablation table currently undercuts the SOTA claim. read the letter →

arxiv 2507.00519 v1 pith:H65EWCRY submitted 2025-07-01 cs.CV

Ruize Cui , Jiaan Zhang , Jialun Pei , Kai Wang , Pheng-Ann Heng , Jing Qin This is my paper · ORCID
classification cs.CV
keywords topology-constrainedlearninglaparoscopicliverlandmarkdetectionpersistenthomologylosscenter-lineconstraintRGB-Dfusiondynamicsnakeconvolutiondepthestimationsurgicalaugmentedreality
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

Liver landmarks are thin, winding anatomical structures inside the abdomen, and this paper's claim is that a detector trained to respect their topology—their connectedness and branching—rather than only their pixel overlap will find them reliably even when instruments, blood, or similar tissue are in the way. To test that, the authors build TopoNet, a two-path network: a ResNet reads the RGB image for texture, while stacking snake-convolution blocks read estimated depth maps for the landmarks' tubular shape. A boundary-aware fusion module merges the two streams, and the training objective adds a center-line constraint and a persistent-homology loss that enforce homotopy equivalence between prediction and ground truth. On the L3D and P2ILF benchmarks the paper reports the best numbers among twelve compared methods, with the largest gain in average symmetric surface distance, and about a 3.4 times faster inference than its closest competitor.

What carries the argument

The load-bearing mechanism is the topological constraint loss, which pairs a multi-class center-line constraint (an extension of clDice) with a persistent-homology loss. The persistence loss computes matched and unmatched connected components between prediction and ground-truth barcodes; matched components are penalized for birth–death interval disagreement, unmatched predicted components are penalized for their persistence length, so predictions are pulled toward homotopy equivalence with labels. The architecture that feeds this loss is the snake-CNN dual-path encoder: STA blocks apply dynamic snake convolution to depth maps to capture tortuous tubular geometry, a ResNet-34 extracts RGB texture, and boundary-aware topological fusion modules combine the streams with attention and a boundary-enhancement residual.

What would settle it

Replace the estimated depth maps in the frozen AdelaiDepth stream with random noise while keeping the same training pipeline; if DSC, IoU, and Assd stay near the reported values, the depth-informed topology claim is not load-bearing, and if the gains disappear the method depends on depth quality. A second check is to measure AdelaiDepth's depth error against the L3D keyframes and see whether frames with poor depth are exactly the frames where TopoNet's Assd advantage shrinks.

Watch

Extended reading notes

Core claim

The paper's central result is that explicit topological supervision changes which errors a landmark detector makes. On L3D, TopoNet improves Dice from 63.52 to 65.19 and IoU from 48.68 to 50.56 over D2GPLand, while average symmetric surface distance drops from 59.38 to 28.07 pixels; on P2ILF the gains are 0.81 in Dice, 1.01 in IoU, and 8.57 pixels in Assd. Inference time falls from 297.93 ms to 86.43 ms and computation from 572.85 to 276.99 GFLOPs. The authors attribute the large Assd improvement to the persistence loss suppressing topological false positives caused by surgical instruments and blood, which sit far from any true landmark pixel.

Load-bearing premise

The depth maps fed to the snake-CNN path come from a frozen monocular depth estimator, and the method's gains assume those depth estimates are accurate and aligned with the RGB frames on the glossy, deformable liver surface.

Editorial extensions

If this is right

  • TopoNet reports the best DSC, IoU, and Assd on both L3D and P2ILF compared with twelve prior methods, including transformer- and SAM-based segmenters.
  • The Assd drop from 59.38 to 28.07 pixels on L3D suggests that far-off false positives from occluding instruments and blood are largely removed rather than merely shrunk.
  • At 86.43 ms per frame and 276.99 GFLOPs, the model is light enough that intraoperative landmark localization for AR-guided liver surgery is feasible on current GPU hardware.
  • The center-line and persistence losses are direct extensions of clDice and persistent homology, so the same training recipe can be applied to other tubular structures without architectural changes.

Reading between the lines

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

  • In my reading, the depth maps are the fragile link: the gains are only as reliable as the frozen AdelaiDepth estimates, and the paper reports no validation of depth quality on the glossy, deforming liver surface; I would expect accuracy to degrade if the depth stream is perturbed.
  • The persistence loss is stated generically, so it could be lifted to vessel, nerve, or duct segmentation tasks where the target topology is known, with the same matched/unmatched barcode penalty replacing pixelwise losses.
  • A direct comparison against an RGB-only TopoNet variant (no STA depth path) would isolate how much of the 1.67-point DSC gain comes from depth geometry versus from the topological losses themselves; the ablations currently remove losses and fusion but not the depth pathway.
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

4 major / 7 minor

Summary. This paper proposes TopoNet, a framework for laparoscopic liver landmark detection that combines a snake-convolution and ResNet dual-path encoder to extract depth-informed topological features and RGB texture features, a boundary-aware topological fusion (BTF) module that merges the two modalities while emphasizing edge regions, and a topological constraint loss comprising a multi-class center-line (clDice-style) loss and a persistent-homology-based matching loss (Eqs. 4-9). On the L3D and P2ILF datasets, the authors report state-of-the-art DSC/IoU/Assd across 12 baselines (Table 1), with the headline result of 65.19 DSC on L3D versus 63.52 for the prior best method D2GPLand, a large reduction of average symmetric surface distance (28.07 vs 59.38), roughly 4x faster inference (86.43 ms vs 297.93 ms), and about half the FLOPs. Ablation studies (Table 2, Fig. 5) attribute gains to the BTF module and to each topological loss component. The central problem identified in this review is that the full model's L3D scores in the ablation table (59.79 DSC, 47.38 IoU, 29.27 Assd) differ from the headline test-set scores (65.19 DSC, 50.56 IoU, 28.07 Assd) without any stated explanation of the evaluation protocol.

Significance. If the reported numbers are reproducible, the paper would make a solid contribution to intraoperative landmark detection: the loss formulations in Eqs. (4)-(9) are concretely specified, the design motivations (snake convolutions for tubular anatomy, depth-informed topology, explicit RGB-D fusion) are sensible, and the efficiency gains over D2GPLand are substantial and clinically relevant. The paper also reports a more complete efficiency profile (inference speed and GFLOPs) than is typical for this literature, and the component ablations are informative as far as they go. However, the unresolved discrepancy between Table 1 and Table 2, the absence of error bars or significance tests on margins of about 1-2 DSC points, and the self-defined random P2ILF split mean that the state-of-the-art claim is not verifiable from the manuscript as written. These issues appear fixable, and the underlying approach remains promising.

major comments (4)
  1. [3.3/Table 1 vs 3.4/Table 2] The full TopoNet model is reported with L3D DSC/IoU/Assd of 65.19/50.56/28.07 in Table 1 but 59.79/47.38/29.27 in Table 2, although Section 3.4 states that the ablations use "the evaluation set of L3D." If "evaluation set" is the same test set used for Table 1, the two tables are mutually contradictory, and the state-of-the-art claim in Section 3.3 fails, since the Table 2 full model would rank below D2GPLand (63.52 DSC). If the ablation instead uses a validation subset or another split, that must be stated explicitly, baseline methods must be re-run on the same split, and the protocol difference must be reconciled; as written, no downstream claim (depth fusion, topological losses, or efficiency) can be evaluated against a fixed comparison protocol.
  2. [3.3 and 3.4] Every metric in Tables 1 and 2 is a single-run point estimate with no error bars, no number of repeated runs, and no statistical significance test. The headline margins over D2GPLand (1.67 DSC on L3D, 0.81 DSC on P2ILF) are smaller than or comparable to the unexplained 5.40-point DSC gap between the two reported evaluations of the full TopoNet model, so run-to-run variability alone could account for the reported gains. Please report mean +/- standard deviation over at least three runs with different seeds and a paired significance test for the TopoNet-versus-D2GPLand comparison on both datasets.
  3. [3.1 and Table 1 (P2ILF)] The P2ILF comparison is conducted on a self-defined random split (124 training images and 43 test images) with no seed stated and no repeated splits, and the paper does not state whether the 12 baseline numbers in the P2ILF column were obtained by re-running those methods under this identical split or quoted from prior publications that used different splits. Unless every baseline was re-trained and evaluated under exactly this split with the same protocol, the P2ILF column of Table 1 is not a controlled comparison, and the reported 0.81 DSC improvement cannot be interpreted; if the baselines were re-run, the training details for each one must be reported.
  4. [2.1, 3.2, and Table 2] The entire depth pathway depends on depth maps produced by the frozen monocular estimator AdelaiDepth, which is not validated on laparoscopic imagery, yet the paper provides no analysis of depth quality or of RGB-depth alignment and no ablation that removes the depth pathway (all Table 2 variants still use depth, with or without BTF). As a concrete test, please report an RGB-only variant (STA blocks removed and the BTF module fed with R_i only) together with a brief depth-quality evaluation, so that the gains attributed to "depth-informed topological structures" can be separated from the mere addition of a second modality and the reader can assess the risk of failure on glossy, deformable liver surfaces.
minor comments (7)
  1. [2.2 and Eq. (3)] The sentence "the other ones utilize only the former fused feature F_{i-1} for residual learning" is ambiguous: it is unclear whether BTF modules 2-5 still receive R_i and D_i as inputs or only F_{i-1}; the figure suggests both modalities remain inputs, so the text should describe the data flow unambiguously.
  2. [3.1] The sentence describing the P2ILF split ("Since only the training set can be available, we randomly select 124 images for training and the remaining 43 images for testing") is confusing; please state clearly that 16 of the 183 images form a held-out set with unavailable labels and that the reported experiments re-split the 167-image training set into 124/43, and report the random seed.
  3. [Fig. 1 caption] The caption says that a broken landmark is highlighted by a yellow circle and two topological false positives by red circles, but it does not state in which panels these circles appear; readers cannot tell whether the broken landmark is in the D2GPLand prediction (c) or elsewhere.
  4. [2.3 and Abstract] The claim that the loss ensures "homotopy equivalence between predictions and labels" is stronger than what Eqs. (6)-(8) implement, which is a 0-dimensional persistent-homology (connected-component) matching loss plus center-line connectivity; please rephrase to describe the actual topological consistency enforced.
  5. [3.4 and Fig. 5] The backbone ablation is presented only as a figure with no numerical values; please provide a table (or add values to the caption) so that the ResNet-34 choice is quantitatively justified, and state which backbone variants were compared.
  6. [3.3] Minor wording and reporting issues: "the inference speed of TopoNet outperform all compared models" should read "outperforms," "GLOPs" should be "GFLOPs," and the paper should state the input resolution used for the speed and GFLOPs measurements.
  7. [1 and Related Work] The paper has no related-work section and does not compare against existing topology-preserving losses for segmentation (e.g., the TopologyLoss of Hu et al., NeurIPS 2019, or other Betti-matching losses); adding such a comparison would clarify the novelty of L_per and help position the contribution.

Circularity Check

0 steps flagged · score 2.0 of 10

No constructional circularity: the central claim is an empirical benchmark result, not a derivation from fitted inputs. The only self-referential elements (citing the authors' own D2GPLand for the depth prior and evaluation protocol) are mild and not load-bearing reductions, though a serious internal table inconsistency independently undermines the SOTA claim.

full rationale

The paper's derivation chain is not circular in the sense of reducing to its own inputs. The proposed method is an empirical pipeline: frozen AdelaiDepth depth maps feed a snake-CNN dual-path encoder, RGB and depth features are fused by the BTF module, and supervised losses (Dice, center-line clDice-style constraint, topological persistence loss) are optimized against ground-truth landmark labels. None of the presented equations (Eqs. 1-9) defines a predicted quantity as a fitted parameter renamed as a discovery; the topological persistence loss directly compares prediction and ground truth barcodes, which is a training objective rather than a self-referential claim. The paper does cite the authors' own prior D2GPLand [16] for two premises: that depth provides useful geometric information, and for the experimental setup/dataset L3D. This is a self-citation, but it refers to a published, externally evaluated method, and the current paper's SOTA claim is an independent empirical comparison on fixed test sets; the depth prior is not 'proved' by the current paper, and the improvement over D2GPLand is testable. Thus there is no exhibited reduction where an output equals an input by construction. However, a notable internal inconsistency exists: Table 1 reports TopoNet at 65.19 DSC / 50.56 IoU / 28.07 Assd on L3D, while Table 2, described as using the L3D evaluation set, reports the full TopoNet at 59.79 DSC / 47.38 IoU / 29.27 Assd. If both tables refer to the same test set, the full model in the ablation would be below D2GPLand's 63.52 DSC, contradicting the SOTA claim; if they refer to different splits, the protocol is not held fixed. This is a correctness/reporting problem, not circularity, and it is why the non-circularity score is not 0. Overall, the paper shows no significant constructional circularity, but the self-referential benchmark framing and the table discrepancy warrant caution.

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

No new physical entities are introduced. The framework relies on standard deep learning components and an external depth estimator, plus a set of empirically chosen hyperparameters. The main unproven inputs are the reliability of the depth maps and the correctness of the persistence-barcode matching.

free parameters (3)
  • Loss balancing weights = lambda_d=0.4, lambda_cl=0.4, lambda_per=0.2
    Set empirically in Sec. 3.2 without a search or sensitivity analysis; they directly affect the training objective and the reported results.
  • Persistence loss smoothing factor = s=1e-5
    Introduced in Eq. 6-7 to avoid division by zero; the paper gives no analysis of its impact.
  • Training hyperparameters = LR 8e-5 to 1e-6, weight decay 3e-5, batch size 4, 100 epochs
    Given in Sec. 3.2; the warmup strategy for the topology losses is described only qualitatively.
assumptions (3)
  • domain assumption Frozen AdelaiDepth produces accurate monocular depth maps for laparoscopic scenes.
    The STA blocks in Sec. 2.1 are built to extract depth-informed topological structures; the paper does not analyze depth quality or failure cases.
  • domain assumption The Betti matching algorithm from [22] correctly matches predicted and ground-truth connected components for the persistence loss.
    Eq. 6-8 depend on matched (M) and unmatched (U) component sets, but the matching criterion is not specified in the paper.
  • domain assumption The L3D and P2ILF datasets, as split in [16], provide unbiased evaluations.
    The evaluation protocol is inherited from the authors' prior work D2GPLand without independent verification of the split or labels.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Topology-Constrained Learning for Efficient Laparoscopic Liver Landmark Detection." pith.science (2026). https://pith.science/paper/H65EWCRY

@misc{pith2026250700519,
  author       = {Pith},
  title        = {Pith review of: Topology-Constrained Learning for Efficient Laparoscopic Liver Landmark Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H65EWCRY}},
  note         = {Machine review of arXiv:2507.00519}
}
read the original abstract

Liver landmarks provide crucial anatomical guidance to the surgeon during laparoscopic liver surgery to minimize surgical risk. However, the tubular structural properties of landmarks and dynamic intraoperative deformations pose significant challenges for automatic landmark detection. In this study, we introduce TopoNet, a novel topology-constrained learning framework for laparoscopic liver landmark detection. Our framework adopts a snake-CNN dual-path encoder to simultaneously capture detailed RGB texture information and depth-informed topological structures. Meanwhile, we propose a boundary-aware topology fusion (BTF) module, which adaptively merges RGB-D features to enhance edge perception while preserving global topology. Additionally, a topological constraint loss function is embedded, which contains a center-line constraint loss and a topological persistence loss to ensure homotopy equivalence between predictions and labels. Extensive experiments on L3D and P2ILF datasets demonstrate that TopoNet achieves outstanding accuracy and computational complexity, highlighting the potential for clinical applications in laparoscopic liver surgery. Our code will be available at https://github.com/cuiruize/TopoNet.

Figures

Figures reproduced from arXiv: 2507.00519 by the authors.

Figure 1
Figure 1. Illustration of the crucial role of topological priors in tubular character [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overall architecture of the proposed TopoNet. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Visual comparison of our TopoNet with representative models. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Visualizations of class-aware attention maps with our predictions. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Ablations for backbones [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 20 canonical work pages

  1. [1]

    MedIA99, 103371 (2025)

    Ali, S., Espinel, Y., Jin, Y., Liu, P., Güttner, B., Zhang, X., Zhang, L., Dowrick, T., Clarkson, M.J., Xiao, S., et al.: An objective comparison of methods for augmented reality in laparoscopic liver resection by preoperative-to-intraoperative image fu- sion from the MICCAI2022 challenge. MedIA99, 103371 (2025)

    work page 2025

  2. [2]

    In: ECCV (2022)

    Cao, H., Wang, Y., Chen, J., Jiang, D., Zhang, X., Tian, Q., Wang, M.: Swin-unet: Unet-like pure transformer for medical image segmentation. In: ECCV (2022)

    work page 2022

  3. [3]

    Procedia Computer Science 246, 4951–4958 (2024)

    Chai, S., Jain, R.K., Teng, S., Liu, J., Li, Y., Tateyama, T., Chen, Y.w.: Lad- der fine-tuning approach for sam integrating complementary network. Procedia Computer Science 246, 4951–4958 (2024)

    work page 2024

  4. [4]

    arXiv preprint arXiv:2102.04306 (2021)

    Chen, J., Lu, Y., Yu, Q., Luo, X., Adeli, E., Wang, Y., Lu, L., Yuille, A.L., Zhou, Y.:Transunet:Transformersmakestrongencodersformedicalimagesegmentation. arXiv preprint arXiv:2102.04306 (2021)

    arXiv 2021

  5. [5]

    In: IEEE ICCV

    Chen, T., Zhu, L., Deng, C., Cao, R., Wang, Y., Zhang, S., Li, Z., Sun, L., Zang, Y., Mao, P.: Sam-adapter: Adapting segment anything in underperformed scenes. In: IEEE ICCV. pp. 3367–3375 (2023)

    work page 2023

  6. [6]

    IEEE TMI 40(1), 371–380 (2020)

    Collins, T., Pizarro, D., Gasparini, S., Bourdel, N., Chauvet, P., Canis, M., Calvet, L., Bartoli, A.: Augmented reality guided laparoscopic surgery of the uterus. IEEE TMI 40(1), 371–380 (2020)

    work page 2020

  7. [7]

    American Mathematical Soc

    Edelsbrunner, H., Harer, J.: Computational topology: an introduction. American Mathematical Soc. (2010)

    2010

  8. [8]

    Annals of surgery267(2), 199–207 (2018)

    Fretland, Å.A., Dagenborg, V.J., Bjørnelv, G.M.W., Kazaryan, A.M., Kristiansen, R., Fagerland, M.W., Hausken, J., Tønnessen, T.I., Abildgaard, A., Barkhatov, L., et al.: Laparoscopic versus open resection for colorectal liver metastases: the oslo-comet randomized controlled trial. Annals of surgery267(2), 199–207 (2018)

    work page 2018

Show all 27 references
  1. [9]

    Surgical Endoscopy pp

    Giannone, F., Cassese, G., Del Basso, C., Alagia, M., Palucci, M., Sangiuolo, F., Panaro, F.: Robotic versus laparoscopic liver resection for difficult posterosuperior segments: a systematic review with a meta-analysis of propensity-score matched studies. Surgical Endoscopy pp...

    2024

  2. [10]

    In: IEEE CVPR

    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: IEEE CVPR. pp. 770–778 (2016)

    2016

  3. [11]

    arXiv preprint arXiv:2306.13731 (2023)

    Hu, X., Xu, X., Shi, Y.: How to efficiently adapt large segmentation model (sam) to medical images. arXiv preprint arXiv:2306.13731 (2023)

    2023 arXiv

  4. [12]

    IEEE TAI (2024)

    Hu, X., Cutolo, F., Iqbal, H., Henckel, J., y Baena, F.R.: Artificial intelligence- driven framework for augmented reality markerless navigation in knee surgery. IEEE TAI (2024)

    2024

  5. [13]

    arXiv preprint arXiv:1809.10486 (2018) 10 R

    Isensee,F.,Petersen,J.,Klein,A.,Zimmerer,D.,Jaeger,P.F.,Kohl,S.,Wasserthal, J., Koehler, G., Norajitra, T., Wirkert, S., et al.: nnu-net: Self-adapting frame- work for u-net-based medical image segmentation. arXiv preprint arXiv:1809.10486 (2018) 10 R. Cui et al

    2018 arXiv

  6. [14]

    In: MIDL (2023)

    Labrunie, M., Pizarro, D., Tilmant, C., Bartoli, A.: Automatic 3d/2d deformable registration in minimally invasive liver resection using a mesh recovery network. In: MIDL (2023)

    2023

  7. [15]

    IJCARS 17(8), 1429–1436 (2022)

    Labrunie, M., Ribeiro, M., Mourthadhoi, F., Tilmant, C., Le Roy, B., Buc, E., Bartoli, A.: Automatic preoperative 3d model registration in laparoscopic liver resection. IJCARS 17(8), 1429–1436 (2022)

    2022

  8. [16]

    In: MICCAI

    Pei, J., Cui, R., Li, Y., Si, W., Qin, J., Heng, P.A.: Depth-driven geometric prompt learning for laparoscopic liver landmark detection. In: MICCAI. pp. 154–164 (2024)

    2024

  9. [17]

    In: IEEE ICCV

    Qi, Y., He, Y., Qi, X., Zhang, Y., Yang, G.: Dynamic snake convolution based on topological geometric constraints for tubular structure segmentation. In: IEEE ICCV. pp. 6070–6079 (2023)

    2023

  10. [18]

    In: MICCAI

    Ronneberger, O., Fischer, P., Brox, T.: U-net: Convolutional networks for biomed- ical image segmentation. In: MICCAI. pp. 234–241 (2015)

    2015

  11. [19]

    Surgical Oncology38, 101637 (2021)

    Schneider, C., Allam, M., Stoyanov, D., Hawkes, D., Gurusamy, K., Davidson, B.: Performance of image guided navigation in laparoscopic liver surgery–a systematic review. Surgical Oncology38, 101637 (2021)

    2021

  12. [20]

    In: IEEE CVPR

    Shit, S., Paetzold, J.C., Sekuboyina, A., Ezhov, I., Unger, A., Zhylka, A., Pluim, J.P., Bauer, U., Menze, B.H.: cldice-a novel topology-preserving loss function for tubular structure segmentation. In: IEEE CVPR. pp. 16560–16569 (2021)

    2021

  13. [21]

    Heliyon 10(3) (2024)

    Smithmaitrie, P., Khaonualsri, M., Sae-Lim, W., Wangkulangkul, P., Jearanai, S., Cheewatanakornkul, S.: Development of deep learning framework for anatomical landmarkdetectionandguideddissectionlineduringlaparoscopiccholecystectomy. Heliyon 10(3) (2024)

    2024

  14. [22]

    Stucki, N., Bürgin, V., Paetzold, J.C., Bauer, U.: Efficient betti matching enables topology-aware 3d segmentation via persistent homology (2024)

    2024

  15. [23]

    IEEE TPAMI43(10), 3349–3364 (2020)

    Wang, J., Sun, K., Cheng, T., Jiang, B., Deng, C., Zhao, Y., Liu, D., Mu, Y., Tan, M., Wang, X., et al.: Deep high-resolution representation learning for visual recognition. IEEE TPAMI43(10), 3349–3364 (2020)

    2020

  16. [24]

    In: ITME

    Xiao, X., Lian, S., Luo, Z., Li, S.: Weighted res-unet for high-quality retina vessel segmentation. In: ITME. pp. 327–331 (2018)

    2018

  17. [25]

    IEEE TPAMI (2022)

    Yin, W., Zhang, J., Wang, O., Niklaus, S., Chen, S., Liu, Y., Shen, C.: Towards accurate reconstruction of 3d scene shape from a single monocular image. IEEE TPAMI (2022)

    2022

  18. [26]

    arXiv preprint arXiv:2304.13785 (2023)

    Zhang, K., Liu, D.: Customized segment anything model for medical image seg- mentation. arXiv preprint arXiv:2304.13785 (2023)

    2023 arXiv

  19. [27]

    IEEE TMI39(6), 1856–1867 (2019)

    Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N., Liang, J.: Unet++: Redesigning skip connections to exploit multiscale features in image segmentation. IEEE TMI39(6), 1856–1867 (2019)

    2019

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.