Pith. sign in

REVIEW 3 major objections 6 minor 1 cited by

COMPrompter: reconceptualized segment anything model with multiprompt network for camouflaged object detection

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Multiprompt SAM variant reaches state-of-the-art camouflaged object detection

desk verdict A clever SAM multiprompt design, but the headline SOTA margin relies on GT-derived prompts at inference and does not survive contact with an automatic comparison. read the letter →

arxiv 2411.18858 v1 pith:TH67IREW submitted 2024-11-28 cs.CV

Xiaoqin Zhang , Zhenni Yu , Li Zhao , Deng-Ping Fan , Guobao Xiao This is my paper · ORCID
classification cs.CV
keywords camouflagedobjectdetectionsegmentanythingmodelmultipromptnetworkboundarypromptedgegradientextractionbox-boundarymutualguidancediscretewavelettransformpolypsegmentation
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 paper tries to show that the Segment Anything Model can be turned into a leading camouflaged object detector by replacing its single box prompt with a multiprompt strategy. The new network, COMPrompter, adds a boundary prompt that encodes edge-gradient information and fuses it with the original box prompt, plus high-frequency features from a discrete wavelet transform. If correct, the result is the best published scores on four COD benchmarks and five polyp segmentation datasets, including an average 2.2% improvement over the prior leading model on COD10K. The claim matters because it points a way to adapt large segmentation foundation models to difficult, fine-structured tasks without retraining their heavy image encoders.

What carries the argument

The load-bearing mechanism is the boundary prompt generated by EGEM, following $BG = \gamma(\mu(GT) - GT) * C(I)$, where $\mu$ is a $3\times3$ dilation, $\gamma$ a $5\times5$ dilation, and $C$ a Canny operation. This gradient-bearing boundary mask is encoded by a frozen prompt encoder and combined with the dense box embedding through BBMG, which uses adapted pointwise depthwise-separable convolution with residual connections; the diagonal high-frequency component $HF = x_1 - x_2 - x_3 + x_4$ from the discrete wavelet transform is then concatenated and convolved to form the final dense embedding that guides the mask decoder.

What would settle it

Run COMPrompter on COD10K with box and boundary prompts generated automatically (for example, from a pretrained detector and the paper's UEDG pipeline with offset 15) instead of from ground truth, and compare it against the same competitors under identical conditions; if the reported margin over the next-best method shrinks below statistical significance or inverts, the state-of-the-art claim does not survive real deployment.

Watch

Extended reading notes

Core claim

COMPrompter establishes that a SAM-based detector can surpass specialized camouflaged object detection methods when the prompt is enriched rather than left as a single box. The paper proposes a boundary prompt built by an edge gradient extraction module that dilates the ground-truth mask, subtracts the original mask, and multiplies the result with a Canny edge map of the image, producing a boundary mask that carries gradient information at the object-background junction. A box-boundary mutual guidance module then fuses dense box and boundary embeddings, and diagonal high-frequency features from a discrete wavelet transform are appended as a supplementary signal. With the image encoder and prompt encoder frozen and only the mask decoder fine-tuned, the method reports state-of-the-art results on CAMO, CHAMELEON, COD10K, and NC4K, and on five polyp segmentation benchmarks, while cutting SAM's parameter count to 94.86M and quadrupling its inference speed.

Load-bearing premise

The benchmark scores assume that ground-truth-derived box and boundary prompts are a fair stand-in for real user prompts, while all automatic comparison methods must detect the camouflaged object without that target-specific information.

Editorial extensions

If this is right

  • Any SAM-based camouflaged object detector could adopt the boundary prompt and expect a large jump: adding the boundary prompt to box-only SAM raises weighted F-measure by 5.8 percentage points on average across COD10K and NC4K.
  • The diagonal high-frequency (HH) component of the DWT is the useful band for COD; it outperforms LL, LH, and HL on all four metrics, so future SAM adaptations can focus on that component.
  • The approach transfers to polyp segmentation, where it reports gains of about 3.2% in mean Dice and 4.5% in mean IoU over MedSAM across the five evaluated datasets.
  • Because the heavy encoders are frozen and precomputed, the method is trainable in about 4.2 hours on a single RTX 3080Ti, suggesting foundation-model fine-tuning for COD is feasible without large compute budgets.

Reading between the lines

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

  • The reported superiority depends on ground-truth-derived prompts at inference; a practical deployment would need an automatic box and boundary generator, so the approach is best read as a prompt-design result rather than a fully autonomous detector.
  • The boundary-prompt mechanism is domain-agnostic and could plausibly transfer to other edge-critical segmentation tasks such as aerial imagery, satellite building extraction, or lesion boundary delineation, provided the dilation kernels and binarization offset are retuned.
  • The paper's own note that a single box can miss multiple targets suggests a natural extension: using several sub-boxes with per-box boundary prompts, which would also make the method more robust to multi-instance scenes.
  • If a paired test were run where competitors also receive the same box and boundary prompts, the comparison would isolate the value of the multiprompt fusion itself; current tables conflate prompt advantage with network advantage.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper proposes COMPrompter, a SAM-based multiprompt network for camouflaged object detection. It introduces a boundary prompt computed by an edge gradient extraction module (EGEM), a box-boundary mutual guidance module (BBMG), and discrete wavelet transform high-frequency features. The model is evaluated on CAMO, CHAMELEON, COD10K, NC4K, and five polyp segmentation datasets, with claims of state-of-the-art performance. The main quantitative results use ground-truth-derived boundary and box prompts at inference; a separate practical variant (Ours*) uses automatically generated boundaries and is reported to rank within the top three.

Significance. If the method were shown to outperform prior automatic methods under identical task conditions, the multiprompt design and frequency-domain augmentation would be of interest to COD and SAM-adaptation research. The internal ablations indicate that each proposed component contributes in the prompted setting, and the precomputation of image embeddings is an efficiency measure worth noting. However, the empirical case for state-of-the-art performance is not established because the comparison conditions differ between the proposed method and the baselines, and because key hyperparameters are tuned on the test benchmarks.

major comments (3)
  1. [Section 3.1, Eq. (1); Table 1] The main comparison in Table 1 does not measure the same task for COMPrompter and the baseline methods. Section 3.1 states that during inference "we use the GT to generate boundaries and boxes as user prompts to simulate scenarios of user interaction," and Eq. (1) defines the boundary prompt as BG = gamma(mu(GT) - GT) * C(I), so the target mask is injected into the model at test time. The automatic COD methods listed in Table 1 receive no such target-derived location or boundary information, and the SAM, SAM-Adapter, and MedSAM rows do not appear to receive the same box and boundary prompts. Consequently, the claimed SOTA margins, including the abstract's 2.2% average positive gain on COD10K, are not supported. The paper's own automatic variant (Ours*, built from UEDG edges plus binarization) is described in Section 4.3 as ranking only "within the top three," which is consistent with this concern. A fair comparison would need either GT-prompted baselines (e.g., SAM and MedSAM given GT boxes and boundaries) or a fully automatic prompt source for all methods.
  2. [Tables 5 and 6; Section 4.4] The dilation kernel sizes (D1-D5) in Table 5 and the binarization offset value in Table 6 are selected by comparing metrics on the same test datasets (COD10K, NC4K, CAMO, CHAMELEON) used for the final claim. This post hoc selection on the test benchmarks means the reported results are partly fitted; the paper does not provide a held-out validation split, error bars, or repeated-run statistics. The ablation in Table 3 therefore demonstrates internal usefulness of the proposed modules only under this test-set-selected configuration and does not quantify the stability or generalization of the reported gains.
  3. [Table 2; Section 3.1] The polyp segmentation comparison in Table 2 is subject to the same GT-prompting issue as the COD comparison. The global inference protocol of Section 3.1 ("we use the GT to generate boundaries and boxes as user prompts") appears to apply to the polyp experiments as well, so the paper does not establish that COMPrompter outperforms the listed polyp methods under an equivalent task definition. The internal ablations (M1 to M5 in Table 3) are informative about the relative contribution of the box prompt, boundary prompt, and DWT once GT prompts are available, but they do not justify the headline SOTA claims.
minor comments (6)
  1. [Section 4.1] The section title "Datatset" is a typo for "Dataset."
  2. [Table 2] The header "CVC-linicDB" is a typo for "CVC-ClinicDB," which is the name used elsewhere in the paper.
  3. [Figure 1] The scatter plot in Figure 1 is difficult to read because the x-axis is MAE and the y-axis is the sum of three metrics with different scales; the caption should state any normalization explicitly, and Figure 1 also contains the typo "T est."
  4. [Abstract and Conclusion] There are several grammatical slips, including "achieves a the cutting-edge performance" in the contributions and "It use a multiprompt strategy" in the Conclusion.
  5. [Table 1] The "Ours*" column is introduced in Table 1 without a table footnote; the explanation of the star appears only later in Section 4.3, which makes the table initially confusing.
  6. [Code availability] The abstract promises code availability, but no code or pretrained weights are provided at this time; please clarify the release status, as this hampers reproducibility of the reported numbers.

Circularity Check

2 steps flagged · score 6.0 of 10

SOTA claim depends on GT-derived box/boundary prompts at inference and test-set-tuned hyperparameters, so the reported margin is not an independent prediction.

  1. self definitional [Section 3.1 (Overall architecture) and Eq. (1) in Section 3.2.1]
    "During inference, we use the GT to generate boundaries and boxes as user prompts to simulate scenarios of user interaction. ... BG = γ(µ(GT) − GT) ∗ C(I)."

    The boundary prompt is literally a function of the target mask: Eq. (1) computes a dilated GT rim multiplied by the image's Canny edges, and the box prompt is also derived from the GT. Thus the 'user prompts' supplied at test time encode the ground-truth object location and boundary. The abstract's 'cutting-edge performance' and the 2.2% COD10K margin in Table 1 are measured for this GT-prompted model against automatic COD methods (UCNet, SINet, MSCAF-Net, etc.) that receive no such oracle prior. The predicted-prompt variant 'Ours*' is acknowledged to rank only 'within the top three,' confirming that the headline gain is not produced by an independent prediction but is in part fed in through the GT-derived input.

  2. fitted input called prediction [Section 4.4, hyperparameter selection (Tables 5 and 6)]
    "Considering potential biases in boundary acquisition during inference, the dilation parameter setting of D1 seemed overly precise ... Therefore, we adopted the parameters from D2. ... As presented in Table 6, the set with an offset of 15 achieved the highest accuracy, with performance decreasing on either side. Therefore, we selected 15 as the optimal offset value."

    The dilation kernels (D1-D5) and the binarization offset are chosen by comparing F_β^ω, S_α, E_ϕ, and M on the same COD10K and NC4K test splits that produce the final reported scores. Selecting the variant with the best test metrics and then presenting those same test metrics as evidence of SOTA is fitting the evaluation target; the reported 'Ours*' prediction is therefore not an out-of-sample result. Because the offset and dilation choice is optimized on the test set, the margin is partly a fitted quantity rather than an independent prediction.

full rationale

The architectural derivation itself is self-contained: EGEM, BBMG, and DWT are defined by explicit equations (1)-(5), and the ablation study shows each component contributes. The circularity is in the evaluation-as-prediction protocol. Section 3.1 states that at inference the GT is used to create both box and boundary prompts, and Eq. (1) makes this explicit: BG = γ(µ(GT) − GT) ∗ C(I), i.e. the boundary prompt is a dilated band around the ground-truth mask. Table 1 then reports this GT-prompted variant ('Ours') as the SOTA result and computes the 2.2% COD10K margin against automatic COD methods that receive no GT-derived prior. The paper's own predicted-prompt variant ('Ours*') is reported to rank only 'within the top three,' so the headline margin depends on oracle prompts. In addition, the offset value of 15 and the dilation kernel D2 are selected by inspecting metrics on the same COD10K and NC4K test sets (Tables 5 and 6), making those reported scores partly fitted. These are prediction-integrity and unfair-comparison issues rather than a formal equivalence between input and output, so the score is 6 rather than 8-10.

Assumptions & free parameters 3 free parameters · 5 assumptions · 3 invented entities

The central claims rest on GT-derived prompts, frozen SAM features, and test-set-tuned hyperparameters. The method introduces no new physical entities, but the boundary prompt, EGEM, and BBMG are internal constructs whose effectiveness is only demonstrated on the paper's own benchmarks.

free parameters (3)
  • Dilation kernel sizes for EGEM (Dilate1=3, Dilate2=5) = 3x3 and 5x5 (setting D2)
    Table 5 compares five dilation settings and selects D2 based on test-set scores on COD10K and NC4K.
  • Binarization offset for generated boundary = 15
    Table 6 compares offsets +5 to +25 and selects +15 based on highest average metrics on the test benchmarks. Affects the Our* results.
  • Canny edge detection thresholds = not specified
    The Canny operator C(I) in Eq. 1 is central to the boundary mask, but its thresholds are not reported; this is an unstated hyperparameter.
assumptions (5)
  • domain assumption Ground-truth-derived box and boundary prompts at inference faithfully simulate real user interactions.
    Used in Section 3.1, where the paper states that GT is used to generate boundaries and boxes as user prompts; this underlies all main Table 1 results.
  • domain assumption Frozen SAM image and prompt encoders preserve sufficient feature quality for COD.
    Section 3.1 freezes the SAM encoders and only fine-tunes the mask decoder, assuming the pretrained features are adequate for camouflage detection.
  • domain assumption High-frequency diagonal DWT components (HH) are the most informative supplement for boundary features.
    Selected via Table 3 and Figure 11 after comparing LL, LH, HL, HH on test sets.
  • standard math Discrete wavelet transform formula HH = x1 - x2 - x3 + x4 is standard.
    Standard 2D Haar wavelet decomposition; no proof is needed.
  • ad hoc to paper Dilation plus Canny multiplication extracts a useful gradient-rich boundary mask.
    Eq. 1 defines the boundary mask BG; its usefulness is only validated on the tested datasets, not grounded in independent evidence.
invented entities (3)
  • Boundary prompt
    purpose: New input prompt carrying edge-gradient information to the frozen SAM prompt encoder.
    No external or theoretical evidence; only internal ablation (M3 vs M1) on COD datasets.
  • Box-boundary mutual guidance module (BBMG)
    purpose: Fuses dense box and boundary embeddings before the mask decoder.
    Internal module; effectiveness shown only through ablations in the same paper.
  • Edge gradient extraction module (EGEM)
    purpose: Generates the gradient boundary mask from GT and image.
    Defined in Section 3.2.1; no external validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of COMPrompter: reconceptualized segment anything model with multiprompt network for camouflaged object detection." pith.science (2026). https://pith.science/paper/TH67IREW

@misc{pith2026241118858,
  author       = {Pith},
  title        = {Pith review of: COMPrompter: reconceptualized segment anything model with multiprompt network for camouflaged object detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TH67IREW}},
  note         = {Machine review of arXiv:2411.18858}
}
read the original abstract

We rethink the segment anything model (SAM) and propose a novel multiprompt network called COMPrompter for camouflaged object detection (COD). SAM has zero-shot generalization ability beyond other models and can provide an ideal framework for COD. Our network aims to enhance the single prompt strategy in SAM to a multiprompt strategy. To achieve this, we propose an edge gradient extraction module, which generates a mask containing gradient information regarding the boundaries of camouflaged objects. This gradient mask is then used as a novel boundary prompt, enhancing the segmentation process. Thereafter, we design a box-boundary mutual guidance module, which fosters more precise and comprehensive feature extraction via mutual guidance between a boundary prompt and a box prompt. This collaboration enhances the model's ability to accurately detect camouflaged objects. Moreover, we employ the discrete wavelet transform to extract high-frequency features from image embeddings. The high-frequency features serve as a supplementary component to the multiprompt system. Finally, our COMPrompter guides the network to achieve enhanced segmentation results, thereby advancing the development of SAM in terms of COD. Experimental results across COD benchmarks demonstrate that COMPrompter achieves a cutting-edge performance, surpassing the current leading model by an average positive metric of 2.2% in COD10K. In the specific application of COD, the experimental results in polyp segmentation show that our model is superior to top-tier methods as well. The code will be made available at https://github.com/guobaoxiao/COMPrompter.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Prompt Engineering in Segment Anything Model: Methodologies, Applications, and Emerging Challenges

    cs.CV 2025-07 conditional novelty 2.0 of 10

    A structured survey of prompt engineering methods for the Segment Anything Model, covering geometric, textual, and multimodal prompts and their applications.

Reference graph

Works this paper leans on

57 extracted references · 52 canonical work pages · cited by 1 Pith paper

  1. [1]

    8 d b0 =

    11em plus .33em minus .07em 4000 4000 100 4000 4000 500 `\.=1000 = #1 \@IEEEnotcompsoconly \@IEEEcompsoconly #1 * [1] 0pt [0pt][0pt] #1 * [1] 0pt [0pt][0pt] #1 * \| ** #1 \@IEEEauthorblockNstyle \@IEEEcompsocnotconfonly \@IEEEauthorblockAstyle \@IEEEcompsocnotconfonly \@IEEEcompsocconfonly \@IEEEauthordefaulttextstyle \@IEEEcompsocnotconfonly \@IEEEauthor...

    arXiv

  2. [2]

    Fan, G.-P

    D.-P. Fan, G.-P. Ji, P. Xu, M.-M. Cheng, C. Sakaridis, and L. Van Gool, ``Advances in deep concealed scene understanding,'' VI, vol. 1, no. 1, p. 16, 2023

    work page 2023

  3. [3]

    Fan, G.-P

    D.-P. Fan, G.-P. Ji, T. Zhou, G. Chen, H. Fu, J. Shen, and L. Shao, ``Pranet: Parallel reverse attention network for polyp segmentation,'' in MICCAI, 2020

    work page 2020

  4. [4]

    Lidbetter, ``Search and rescue in the face of uncertain threats,'' EJOR, vol

    T. Lidbetter, ``Search and rescue in the face of uncertain threats,'' EJOR, vol. 285, no. 3, pp. 1153--1160, 2020

    work page 2020

  5. [5]

    Y. Liu, H. Li, J. Cheng, and X. Chen, ``Mscaf-net: a general framework for camouflaged object detection via learning multi-scale context-aware features,'' IEEE TCSVT, vol. 33, no. 9, pp. 4934--4947, 2023

    work page 2023

  6. [6]

    Y. Sun, G. Chen, T. Zhou, Y. Zhang, and N. Liu, ``Context-aware cross-level fusion network for camouflaged object detection,'' arXiv preprint arXiv:2105.12555, 2021

    arXiv 2021

  7. [7]

    A. Li, J. Zhang, Y. Lv, B. Liu, T. Zhang, and Y. Dai, ``Uncertainty-aware joint salient object and camouflaged object detection,'' in CVPR, 2021

    work page 2021

  8. [8]

    Q. Zhai, X. Li, F. Yang, C. Chen, H. Cheng, and D.-P. Fan, ``Mutual graph learning for camouflaged object detection,'' in CVPR, 2021

    work page 2021

Show all 57 references
  1. [9]

    J. Zhu, X. Zhang, S. Zhang, and J. Liu, ``Inferring camouflaged objects by texture-aware interactive guidance network,'' in AAAI, 2021

    2021

  2. [10]

    Ji, D.-P

    G.-P. Ji, D.-P. Fan, Y.-C. Chou, D. Dai, A. Liniger, and L. Van Gool, ``Deep gradient learning for efficient camouflaged object detection,'' MIR, vol. 20, no. 1, pp. 92--108, 2023

    2023

  3. [11]

    Y. Pang, X. Zhao, T.-Z. Xiang, L. Zhang, and H. Lu, ``Zoom in and out: A mixed-scale triplet network for camouflaged object detection,'' in CVPR, 2022

    2022

  4. [12]

    Y. Pang, X. Zhao, T.-Z. Xiang, L. Zhang, and H. Lu , ``Zoomnext: A unified collaborative pyramid network for camouflaged object detection,'' IEEE TPAMI, pp. 1--16, 2024

    2024

  5. [13]

    Q. Jia, S. Yao, Y. Liu, X. Fan, R. Liu, and Z. Luo, ``Segment, magnify and reiterate: Detecting camouflaged objects the hard way,'' in CVPR, 2022

    2022

  6. [14]

    Fan, G.-P

    D.-P. Fan, G.-P. Ji, M.-M. Cheng, and L. Shao, ``Concealed object detection,'' IEEE TPAMI, vol. 44, no. 10, pp. 6024--6042, 2021

    2021

  7. [15]

    Y. Lv, J. Zhang, Y. Dai, A. Li, B. Liu, N. Barnes, and D.-P. Fan, ``Simultaneously localize, segment and rank the camouflaged objects,'' in CVPR, 2021

    2021

  8. [16]

    Mei, G.-P

    H. Mei, G.-P. Ji, Z. Wei, X. Yang, X. Wei, and D.-P. Fan, ``Camouflaged object segmentation with distraction mining,'' in CVPR, 2021

    2021

  9. [17]

    Fan, G.-P

    D.-P. Fan, G.-P. Ji, G. Sun, M.-M. Cheng, J. Shen, and L. Shao, ``Camouflaged object detection,'' in CVPR, 2020

    2020

  10. [18]

    F. Yang, Q. Zhai, X. Li, R. Huang, A. Luo, H. Cheng, and D.-P. Fan, ``Uncertainty-guided transformer reasoning for camouflaged object detection,'' in ICCV, 2021

    2021

  11. [19]

    Zhang, D.-P

    J. Zhang, D.-P. Fan, Y. Dai, S. Anwar, F. S. Saleh, T. Zhang, and N. Barnes, ``Uc-net: Uncertainty inspired rgb-d saliency detection via conditional variational autoencoders,'' in CVPR, 2020

    2020

  12. [20]

    T. Chen, L. Zhu, C. Ding, R. Cao, S. Zhang, Y. Wang, Z. Li, L. Sun, P. Mao, and Y. Zang, ``Sam fails to segment anything?--sam-adapter: Adapting sam in underperformed scenes: Camouflage, shadow, and more,'' arXiv preprint arXiv:2304.09148, 2023

    2023 arXiv

  13. [21]

    Ma and B

    J. Ma and B. Wang, ``Segment anything in medical images,'' arXiv preprint arXiv:2304.12306, 2023

    2023 arXiv

  14. [22]

    Kirillov, E

    A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y. Lo, P. Dollar, and R. Girshick, ``Segment anything,'' in ICCV, 2023

    2023

  15. [23]

    Sifre, ``Rigid-motion scattering for image classification,'' Ph.D

    L. Sifre, ``Rigid-motion scattering for image classification,'' Ph.D. dissertation, Ecole Polytechnique, CMAP, 2014

    2014

  16. [24]

    C. He, K. Li, Y. Zhang, L. Tang, Y. Zhang, Z. Guo, and X. Li, ``Camouflaged object detection with feature decomposition and edge reconstruction,'' in CVPR, 2023

    2023

  17. [25]

    segment anything

    G.-P. Ji, D.-P. Fan, P. Xu, M.-M. Cheng, B. Zhou, and L. Van Gool, ``Sam struggles in concealed scenes--empirical study on" segment anything",'' arXiv preprint arXiv:2304.06022, 2023

    2023 arXiv

  18. [26]

    S. Zhu, W. Jing, P. Kang, M. Emam, and C. Li, ``Data augmentation and few-shot change detection in forest remote sensing,'' IEEE STARS, vol. 16, pp. 5919--5934, 2023

    2023

  19. [27]

    C. Peng, M. Zhu, H. Ren, and M. Emam, ``Small object detection method based on weighted feature fusion and csma attention module,'' Electronics, vol. 11, no. 16, p. 2546, 2022

    2022

  20. [28]

    Jiang, G

    Y. Jiang, G. Yin, W. Jing, L. Mohaisen, M. Emam, and Y. Yuan, ``Box-spoof attack against single object tracking,'' Applied Intelligence, pp. 1--17, 2024

    2024

  21. [29]

    Y. Lv, J. Zhang, Y. Dai, A. Li, N. Barnes, and D.-P. Fan, ``Toward deeper understanding of camouflaged object detection,'' IEEE TCSVT, vol. 33, no. 7, pp. 3462--3476, 2023

    2023

  22. [30]

    Li, C.-M

    H. Li, C.-M. Feng, Y. Xu, T. Zhou, L. Yao, and X. Chang, ``Zero-shot camouflaged object detection,'' IEEE TIP, vol. 32, pp. 5126--5137, 2023

    2023

  23. [31]

    Zhang, H

    C. Zhang, H. Bi, T.-Z. Xiang, R. Wu, J. Tong, and X. Wang, ``Collaborative camouflaged object detection: A large-scale dataset and benchmark,'' IEEE TNNLS, pp. 1--15, 2023

    2023

  24. [32]

    Huang, H

    Z. Huang, H. Dai, T.-Z. Xiang, S. Wang, H.-X. Chen, J. Qin, and H. Xiong, ``Feature shrinkage pyramid for camouflaged object detection with transformers,'' in CVPR, 2023

    2023

  25. [33]

    Z. Chen, R. Gao, T.-Z. Xiang, and F. Lin, ``Diffusion model for camouflaged object detection,'' in ECAI, 2023

    2023

  26. [34]

    H. Zhu, P. Li, H. Xie, X. Yan, D. Liang, D. Chen, M. Wei, and J. Qin, ``I can find you! boundary-guided separated attention network for camouflaged object detection,'' in AAAI, 2022

    2022

  27. [35]

    G.-P. Ji, L. Zhu, M. Zhuge, and K. Fu, ``Fast camouflaged object detection via edge-based reversible re-calibration network,'' PR, vol. 123, p. 108414, 2022

    2022

  28. [36]

    Y. Sun, S. Wang, C. Chen, and T.-Z. Xiang, ``Boundary-guided camouflaged object detection,'' 2022

    2022

  29. [37]

    Y. Lyu, H. Zhang, Y. Li, H. Liu, Y. Yang, and D. Yuan, ``Uedg:uncertainty-edge dual guided camouflage object detection,'' IEEE TMM, vol. 26, pp. 4050--4060, 2024

    2024

  30. [38]

    D. Sun, S. Jiang, and L. Qi, ``Edge-aware mirror network for camouflaged object detection,'' in ICME. 1em plus 0.5em minus 0.4em IEEE, 2023

    2023

  31. [39]

    B. Dong, J. Pei, R. Gao, T.-Z. Xiang, S. Wang, and H. Xiong, ``A unified query-based paradigm for camouflaged instance segmentation,'' in ACM MM, 2023

    2023

  32. [40]

    Premachandran, B

    V. Premachandran, B. Bonev, X. Lian, and A. Yuille, ``Pascal boundaries: A semantic boundary dataset with a deep semantic boundary detector,'' in WACV, 2017

    2017

  33. [41]

    J. Wang, Z. Wu, J. Chen, X. Han, A. Shrivastava, S.-N. Lim, and Y.-G. Jiang, ``Objectformer for image manipulation detection and localization,'' in CVPR, 2022

    2022

  34. [42]

    W. Liu, X. Shen, C.-M. Pun, and X. Cun, ``Explicit visual prompting for low-level structure segmentations,'' in CVPR, 2023

    2023

  35. [43]

    Ronneberger, P

    O. Ronneberger, P. Fischer, and T. Brox, ``U-net: Convolutional networks for biomedical image segmentation,'' in MICCAI, 2015

    2015

  36. [44]

    Z. Zhou, M. M. Rahman Siddiquee, N. Tajbakhsh, and J. Liang, ``Unet++: A nested u-net architecture for medical image segmentation,'' in MICCAI, 2018

    2018

  37. [45]

    D. Jha, P. H. Smedsrud, M. A. Riegler, D. Johansen, T. De Lange, P. Halvorsen, and H. D. Johansen, ``Resunet++: An advanced architecture for medical image segmentation,'' in ISM, 2019

    2019

  38. [46]

    Y. Fang, C. Chen, Y. Yuan, and K.-y. Tong, ``Selective feature aggregation network with area-boundary constraints for polyp segmentation,'' in MICCAI, 2019

    2019

  39. [47]

    Patel, A

    K. Patel, A. M. Bur, and G. Wang, ``Enhanced u-net: A feature enhancement network for polyp segmentation,'' in CRV, 2021

    2021

  40. [48]

    J. Wei, Y. Hu, R. Zhang, Z. Li, S. K. Zhou, and S. Cui, ``Shallow attention network for polyp segmentation,'' in MICCAI, 2021

    2021

  41. [49]

    Zhang, P

    R. Zhang, P. Lai, X. Wan, D.-J. Fan, F. Gao, X.-J. Wu, and G. Li, ``Lesion-aware dynamic kernel for polyp segmentation,'' in MICCAI, 2022

    2022

  42. [50]

    T. Zhou, Y. Zhou, C. Gong, J. Yang, and Y. Zhang, ``Feature aggregation and propagation network for camouflaged object detection,'' IEEE TIP, vol. 31, pp. 7036--7047, 2022

    2022

  43. [51]

    T.-N. Le, T. V. Nguyen, Z. Nie, M.-T. Tran, and A. Sugimoto, ``Anabranch network for camouflaged object segmentation,'' CVIU, vol. 184, pp. 45--56, 2019

    2019

  44. [52]

    Skurowski, H

    P. Skurowski, H. Abdulameer, J. B aszczyk, T. Depta, A. Kornacki, and P. Kozie , ``Animal camouflage analysis: Chameleon database,'' Unpublished manuscript, vol. 2, no. 6, p. 7, 2018

    2018

  45. [53]

    Silva, A

    J. Silva, A. Histace, O. Romain, X. Dray, and B. Granado, ``Toward embedded detection of polyps in wce images for early diagnosis of colorectal cancer,'' IJCARS, vol. 9, no. 2, pp. 283--293, 2014

    2014

  46. [54]

    Bernal, F

    J. Bernal, F. J. S \'a nchez, G. Fern \'a ndez-Esparrach, D. Gil, C. Rodr \' guez, and F. Vilari \ n o, ``Wm-dova maps for accurate polyp highlighting in colonoscopy: Validation vs. saliency maps from physicians,'' CMIG, vol. 43, pp. 99--111, 2015

    2015

  47. [55]

    Tajbakhsh, S

    N. Tajbakhsh, S. R. Gurudu, and J. Liang, ``Automated polyp detection in colonoscopy videos using shape and context information,'' IEEE TMI, vol. 35, no. 2, pp. 630--644, 2015

    2015

  48. [56]

    V \'a zquez, J

    D. V \'a zquez, J. Bernal, F. J. S \'a nchez, G. Fern \'a ndez-Esparrach, A. M. L \'o pez, A. Romero, M. Drozdzal, A. Courville et al., ``A benchmark for endoluminal scene segmentation of colonoscopy images,'' JHE, vol. 2017, p. 4037190, 2017

    2017

  49. [57]

    D. Jha, P. H. Smedsrud, M. A. Riegler, P. Halvorsen, T. de Lange, D. Johansen, and H. D. Johansen, ``Kvasir-seg: A segmented polyp dataset,'' in MMM, 2020

    2020

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.