Pith. sign in

REVIEW 3 major objections 7 minor 4 cited by

SPT: Sequence Prompt Transformer for Interactive Image Segmentation

T0 review · 3 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Feeding earlier images, clicks, and predicted masks into a causally masked transformer improves interactive segmentation, the paper claims, beating state-of-the-art click methods on five benchmarks.

desk verdict A plausible new sequence-prompt formulation for interactive segmentation, but zero reported numbers make the central claim unverifiable. read the letter →

arxiv 2412.10224 v1 pith:FNU6SEPE submitted 2024-12-13 cs.CV

Senlin Cheng , Haopeng Sun This is my paper · ORCID
classification cs.CV
keywords interactiveimagesegmentationsequencepromptingclick-basedcausalself-attentionpromptselectionDINOv2featuresADE20K-Seqtransformer
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

Interactive image segmentation usually treats each image as an isolated problem: a user clicks, the model segments, repeat per image. This paper claims that when the same object category appears across many images, the history itself is a prompt—the model should read earlier images, clicks, and predicted masks as a sequence and let the current image borrow from similar earlier frames. To make that idea work, the authors build a Sequence Prompt Transformer whose concealed (causal) self-attention hides future frames, and a Top-k Prompt Selection module that uses DINOv2 image features to choose the most similar earlier images as prompts. They report that this setup beats state-of-the-art click-based methods on GrabCut, Berkeley, COCO-MVal, DAVIS, and their new ADE20K-Seq benchmark, with the largest gains at few clicks. If the claim holds, annotating a batch of same-category images would need many fewer clicks, directly cutting the cost of pixel-level data labeling.

What carries the argument

The load-bearing mechanism is the Multi-head Concealed Self-Attention inside the Sequence Prompt Transformer: for a feature sequence $F$, position $i=(x,y)$ in the attention is visible only when $x \ge y$, encoded by a mask function $mask(x,y)=1$ if $x\ge y$ and $0$ otherwise, so each token attends to preceding positions and itself but never to future frames. Each input frame is formed by concatenating the click map and mask, embedding that concatenation, and adding it to the embedded image before the ViT computes $F_i = \mathrm{ViT}(\mathrm{Embed}(C_i \oplus M_i) + \mathrm{Embed}(I_i))$. The Top-k Prompt Selection module supplies the prompt subset by ranking DINOv2 features for similarity to the test image; the SPT output then goes through a Feature Pyramid Module and an MLP Segmentation Head, trained with focal loss.

What would settle it

Re-run SPT on ADE20K-Seq with the TPS module removed (e.g., using the k most recent images as prompts instead of the DINOv2-selected ones) and compare NoC85/NoC90 and M-IoU; if the DINOv2 selection margin disappears, reverses, or shrinks to statistical noise, the claim that Top-k Prompt Selection carries the improvement is false.

Watch

Extended reading notes

Core claim

On its own terms, the paper's discovery is that the sequence of user interactions plus the model's own earlier masks is usable signal, not noise: a causally masked transformer layer placed after a ViT backbone lets the feature of the current image attend to features of earlier images, and the attention is guided by a position-wise mask so no future information leaks. The paper further claims that which earlier images you show matters, and that selecting the top-k most DINOv2-similar images as prompts outperforms using all or the most recent ones. The reported consequence is lower NoC85 and NoC90 on five benchmarks, including the new ADE20K-Seq dataset built from ADE20K sequences of seven categories, and better M-IoU at every click count, especially with very few clicks.

Load-bearing premise

The method's load-bearing premise is that DINOv2 feature similarity between images is a reliable proxy for how useful an earlier image, click, and mask will be as a prompt for the current image; the paper's ablation for this premise is referenced as ExpID #6–#9 but the table itself is missing from the manuscript.

Editorial extensions

If this is right

  • On the paper's reported numbers, a user labeling a series of same-category images should need fewer clicks per image: NoC85 and NoC90 drop relative to RITM, FocalClick, SimpleClick, SAM, HQ-SAM, and the other baselines.
  • Because the gain is largest at one or two clicks (the M-IoU curves in Figure 4), the method is most valuable in the low-interaction regime where single-image models fail hardest.
  • Longer prompt sequences improve accuracy up to the tested length of ten prompts, so practitioners can trade memory for precision by increasing sequence length.
  • The TPS module transfers across datasets: it is trained on COCO and LVIS and evaluated on GrabCut, Berkeley, COCO-MVal, DAVIS, and ADE20K-Seq, so similarity-based prompt retrieval does not need dataset-specific retraining.
  • The new ADE20K-Seq benchmark gives the community a fixed seven-category sequence test set on which future sequence-aware interactive segmentation methods can be compared.

Reading between the lines

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

  • Editorial inference: the success of DINOv2-based TPS points to a testable refinement—train a small retrieval network with the segmentation loss so the prompt selector and segmenter are optimized jointly; the paper does not attempt this.
  • Editorial inference: because DAVIS is a video benchmark, a natural stress test is to compare SPT against video object segmentation methods that also propagate masks over time; the paper only compares against click-based single-image methods, so it has not yet isolated the contribution of sequence prompting from generic mask propagation.
  • Editorial inference: ADE20K-Seq groups static ADE20K images by category rather than by true temporal continuity; a benchmark built from consecutive video frames would test whether the method's gains persist when appearance changes are large and motion blur occurs.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. The paper proposes SPT (Sequence Prompt Transformer), a method for interactive image segmentation that, unlike prior single-image methods, exploits a sequence of images depicting the same object category by using previous images, clicks, and predicted masks as prompts. A Top-k Prompt Selection (TPS) module, based on DINOv2 feature similarity, chooses the most relevant prompts from the sequence. The authors also introduce a new benchmark, ADE20K-Seq, constructed from ADE20K by grouping images into seven categories. The empirical section claims state-of-the-art results on GrabCut, Berkeley, COCO-MVal, DAVIS, and ADE20K-Seq, and ablations are said to validate the contributions of SPT and TPS. However, the manuscript as submitted contains no quantitative results: the main comparison table and the ablation table are both unresolved 'Table ??' placeholders, and the only numerical figure (Fig. 4) is not accompanied by aggregate metrics or a statistical description. The central claim of surpassing prior methods is therefore unverifiable from the provided text.

Significance. If the claimed results were supplied and verified, the paper would address a genuinely new and practically relevant variant of interactive segmentation—segmenting the same object category across a sequence of images—and the proposed architecture (causal-mask Transformer over a sequence of image-click-mask features, plus similarity-based prompt selection) is a plausible design. The introduction of a dedicated sequential interactive-segmentation benchmark is also a useful contribution, provided the dataset is described precisely and released. However, at present the paper contains no quantitative evidence: no NoC85/NoC90, IoU, or MIoU numbers appear anywhere, and the ablation analysis that would isolate the effect of the two core modules is missing. The significance of the work cannot be assessed until these results are actually reported. The idea itself is interesting, but the manuscript in its current form is an incomplete research report rather than a verifiable technical contribution.

major comments (3)
  1. [IV-B, Table ??] The main experimental results are entirely absent. Section IV-B states that 'Our SPT framework consistently outperforms the baselines across all datasets and evaluation metrics' but refers only to an unresolved 'Table ??'. No NoC85, NoC90, IoU, or MIoU values are provided for any dataset or baseline in the entire manuscript. Since the paper's central claim is an empirical superiority claim, the absence of the main comparison table means that claim is wholly unsupported. This is a load-bearing deficiency that must be fixed by reporting the actual numbers, not just a promise of a table.
  2. [IV-C, Table ??] The ablation study is also missing. Section IV-C claims that the Sequence Prompt Transformer improves performance (ExpID #1 vs #5) and that Top-k Prompt Selection is effective (ExpID #6 vs #9), but all of these assertions refer to 'Table ??', which does not appear in the manuscript. Without the ablation table, the attribution of performance gains to SPT and TPS is unfounded, and the paper's two named contributions cannot be independently evaluated. The authors must provide the full ablation results, including the settings for different prompt lengths and different selection methods, before the claims can be taken seriously.
  3. [IV-A.2, ADE20K-Seq] The newly introduced ADE20K-Seq benchmark is described in only two sentences: it 'extend[s] ADE20K dataset into 7 category-specific benchmarks, with each category containing more than 100 images', and it is said to contain random tasks. No details are given on how images are selected, how objects/instances are paired across images, how segmentation masks are obtained, or how the evaluation protocol is defined. Moreover, the dataset is not released. As a result, the reported evaluation on ADE20K-Seq (if any) would be impossible to reproduce, and the benchmark itself is not a usable contribution. This is a major reproducibility gap that should be addressed, for example by describing the construction procedure precisely and providing a public release link or a clear statement of availability.
minor comments (7)
  1. [Abstract and Section I] The abstract says the method segments 'a series of images featuring the same target object', while Section III-A and the ADE20K-Seq description refer to 'same category'. These are different notions: same object identity versus same object class. The paper should clarify which setting is actually addressed and be consistent throughout.
  2. [IV-A, Dataset list] The dataset list in Section IV-A.1 misspells 'LVIS' as 'LIVIS' and writes 'DA VIS' instead of 'DAVIS'. Please correct these typos for the camera-ready version.
  3. [IV-B, Figure 4] Section IV-B refers to 'ADE20K-Sep' (a typo for ADE20K-Seq) and states that Fig. 4 shows MIoU versus number of clicks. However, the figure is not described in any quantitative way: no exact MIoU values, no error bars, and no statistical significance test are reported. As a qualitative plot, it cannot substitute for a numerical comparison table.
  4. [III-D, Equation (4)] The definition of the causal mask function in Eq. (4) is notational unclear: it says 'for an element at position i = (x, y) in the sequence' and then defines mask(x, y), but the roles of x and y (sequence positions versus spatial positions) are not explained. Clarify that x indexes the current sequence element and y indexes the attended element, or rewrite the equation to avoid confusion.
  5. [Fig. 3 caption] The caption of Fig. 3 lists '(e) Results of RITM' and then '(e) Results of SPT (ours)', with the letter (e) repeated. Also, the figure's per-panel IoU values are not accompanied by an aggregate measure over the dataset, and the qualitative selection is not described as representative. Please fix the caption and provide aggregate numbers in the text or table.
  6. [References] References [3] and [46] are duplicated; both are the CLIP paper (Radford et al., 2021). Consolidate them into a single entry.
  7. [Related Work] Related work mentions video-based interactive segmentation methods [47], [48] as inapplicable because of timing/scene inconsistency, but it does not discuss context-aware or memory-based segmentation models such as SegGPT [42] or other 'segment everything in context' approaches. A brief comparison would help position the contribution more accurately.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: central claims are empirical and rest on external benchmarks and pretrained features; unresolved tables are an evidence gap, not a circular derivation.

full rationale

I walked the derivation chain of SPT/TPS. The method takes sequence inputs, extracts features with a ViT backbone initialized from MAE, applies a causal self-attention mask (explicitly defined to prevent future leakage), and is trained on COCO and LVIS with focal loss; no prediction in the paper is obtained by inverting its own inputs. The TPS module uses DINOv2, an externally pretrained model, to rank prompt similarity; this is an empirical design choice, not a quantity fitted to the reported outcome, and the paper's own Table ?? (the ablation) is absent, so the claimed TPS gain is unverifiable rather than definitionally forced. Evaluation is against external interactive-segmentation benchmarks (GrabCut, Berkeley, COCO-MVal, DAVIS); the self-constructed ADE20K-Seq is used only for testing, not for fitting the model or selecting hyperparameters in a way that would close a loop. The unresolved 'Table ??' placeholders for the main comparison and ablation mean the central SOTA claim lacks reported evidence, and the un-released ADE20K-Seq hampers independent replication, but missing data and unverified claims are correctness/completeness problems, not circularity: no equation, fitted parameter, or self-citation chain makes the conclusion equivalent to its premises. I therefore find no significant circularity and score 0.

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

The paper depends on standard deep learning assumptions and a few task-specific assumptions. The main free parameters are architecture and training choices, none of which are justified with reported ablations because the ablations are missing. No new physical or mathematical entities are postulated.

free parameters (5)
  • Top-k value k in TPS
    Number of prompt images selected; chosen by hand, but k is never given a numeric value in the manuscript.
  • Prompt sequence length = 10
    Section IV-A.3 sets prompt sequence length to 10; this is a design choice affecting all experiments.
  • Focal loss alpha and gamma
    Section III-E defines FocalLoss but does not report alpha or gamma values.
  • Number of SPT layers N
    The SPT is said to have N Multi-head Concealed Self-Attention layers, but N is not specified.
  • ViT backbone size = ViT-B / ViT-H
    Section IV-A.3 uses ViT-B and ViT-H with MAE weights; the choice affects results.
assumptions (5)
  • domain assumption Images of the same category share sufficiently similar segmentation structure that prompts from earlier images help segment later images.
    Stated in the Introduction: 'sequence images of the same category, along with their clicks and predicted masks, contain similar information.' This is the core motivation.
  • domain assumption DINOv2 feature similarity is a reliable measure of prompt usefulness.
    Section III-C uses DINOv2 to select the k most similar images as prompts; the supporting ablation is in a missing table.
  • domain assumption The click simulation strategy from RITM accurately mimics real user interactions.
    Section III-A says the sequence is built with the click simulation strategy from RITM; this assumption is imported without validation in this paper.
  • domain assumption Training on COCO and LVIS generalizes to the test benchmarks, including the new ADE20K-Seq.
    Section IV-A.1 states training on COCO and LVIS; evaluation on GrabCut, Berkeley, COCO-MVal, DAVIS, and ADE20K-Seq relies on transfer, which is standard but unstated as an assumption.
  • standard math Causal masking prevents future information leakage and is sufficient to model sequence order.
    The mask function in Section III-D is standard causal attention; this is a standard technique.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SPT: Sequence Prompt Transformer for Interactive Image Segmentation." pith.science (2026). https://pith.science/paper/FNU6SEPE

@misc{pith2026241210224,
  author       = {Pith},
  title        = {Pith review of: SPT: Sequence Prompt Transformer for Interactive Image Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FNU6SEPE}},
  note         = {Machine review of arXiv:2412.10224}
}
read the original abstract

Interactive segmentation aims to extract objects of interest from an image based on user-provided clicks. In real-world applications, there is often a need to segment a series of images featuring the same target object. However, existing methods typically process one image at a time, failing to consider the sequential nature of the images. To overcome this limitation, we propose a novel method called Sequence Prompt Transformer (SPT), the first to utilize sequential image information for interactive segmentation. Our model comprises two key components: (1) Sequence Prompt Transformer (SPT) for acquiring information from sequence of images, clicks and masks to improve accurate. (2) Top-k Prompt Selection (TPS) selects precise prompts for SPT to further enhance the segmentation effect. Additionally, we create the ADE20K-Seq benchmark to better evaluate model performance. We evaluate our approach on multiple benchmark datasets and show that our model surpasses state-of-the-art methods across all datasets.

Figures

Figures reproduced from arXiv: 2412.10224 by the authors.

Figure 1
Figure 1. Segmentation of car windows: (a) Existing methods process individual [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 3
Figure 3. Qualitative analysis on the ADE20K-Sep dataset. (a) Image. [PITH_FULL_IMAGE:figures/full_fig_p002_3.png] view at source ↗
Figure 4
Figure 4. Comparison of MIoU performance with different numbers of clicks [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

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

  1. STeInFormer: Spatial-Temporal Interaction Transformer Architecture for Remote Sensing Change Detection

    cs.CV 2024-12 conditional novelty 6.0 of 10

    STeInFormer enhances remote sensing change detection by interacting bi-temporal features during feature extraction and using fixed DCT frequency components as a parameter-light token mixer.

  2. A Novel Scene Coupling Semantic Mask Network for Remote Sensing Image Segmentation

    eess.IV 2025-01 conditional novelty 5.0 of 10

    SCSM, a scene coupling and semantic mask attention decoder, reports higher accuracy than prior methods on four remote sensing segmentation benchmarks with lower computational cost.

  3. Multi-View Factorizing and Disentangling: A Novel Framework for Incomplete Multi-View Multi-Label Classification

    cs.CV 2025-01 conditional novelty 5.0 of 10

    A two-stage framework that factorizes multi-view representations into view-consistent and view-specific parts, using masked reconstruction, semantic contrast, and a graph disentangling loss, outperforms ten baselines ...

  4. Ultra-High Resolution Segmentation via Boundary-Enhanced Patch-Merging Transformer

    cs.CV 2024-12 conditional novelty 5.0 of 10

    A patch-merging transformer with a boundary-enhanced module improves state-of-the-art ultra-high resolution segmentation accuracy across five benchmarks with comparable memory use.

Reference graph

Works this paper leans on

50 extracted references · 22 canonical work pages · cited by 4 Pith papers

  1. [46]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark et al., “Learning transferable visual models from natural language supervision,” in International conference on machine learning . PMLR, 2021, pp. 8748–8763

    2021

  2. [1]

    Reviving iterative training with mask guidance for interactive segmentation,

    K. Sofiiuk, I. A. Petrov, and A. Konushin, “Reviving iterative training with mask guidance for interactive segmentation,” in 2022 IEEE Inter- national Conference on Image Processing (ICIP) . IEEE, 2022, pp. 3141–3145

    work page 2022

  3. [2]

    Focal loss for dense object detection,

    T.-Y . Lin, P. Goyal, R. Girshick, K. He, and P. Doll ´ar, “Focal loss for dense object detection,” in Proceedings of the IEEE international conference on computer vision , 2017, pp. 2980–2988

    2017

  4. [4]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 770–778

    2016

  5. [5]

    Exploring plain vision transformer backbones for object detection,

    Y . Li, H. Mao, R. Girshick, and K. He, “Exploring plain vision transformer backbones for object detection,” in European conference on computer vision . Springer, 2022, pp. 280–296

    2022

  6. [6]

    On certain integrals of lipschitz- hankel type involving products of bessel functions,

    G. Eason, B. Noble, and I. N. Sneddon, “On certain integrals of lipschitz- hankel type involving products of bessel functions,” Philosophical Transactions of the Royal Society of London. Series A, Mathematical and Physical Sciences , vol. 247, no. 935, pp. 529–551, 1955

    work page 1955

  7. [7]

    Bidirectional recurrent neural net- works,

    M. Schuster and K. K. Paliwal, “Bidirectional recurrent neural net- works,” IEEE transactions on Signal Processing , vol. 45, no. 11, pp. 2673–2681, 1997

    1997

  8. [8]

    Simpleclick: Interactive image segmentation with simple vision transformers,

    Q. Liu, Z. Xu, G. Bertasius, and M. Niethammer, “Simpleclick: Interactive image segmentation with simple vision transformers,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 22 290–22 300

    work page 2023

Show all 50 references
  1. [9]

    Focalclick: Towards practical interactive image segmentation,

    X. Chen, Z. Zhao, Y . Zhang, M. Duan, D. Qi, and H. Zhao, “Focalclick: Towards practical interactive image segmentation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 1300–1309

    2022

  2. [10]

    Empirical evaluation of gated recurrent neural networks on sequence modeling,

    J. Chung, C. Gulcehre, K. Cho, and Y . Bengio, “Empirical evaluation of gated recurrent neural networks on sequence modeling,” arXiv preprint arXiv:1412.3555, 2014

    2014 arXiv

  3. [11]

    Segment anything,

    A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y . Loet al., “Segment anything,” arXiv preprint arXiv:2304.02643 , 2023

    2023 arXiv

  4. [12]

    Long short-term memory,

    S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural computation, vol. 9, no. 8, pp. 1735–1780, 1997

    1997

  5. [13]

    Attention is all you need,

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” Advances in neural information processing systems , vol. 30, 2017

    2017

  6. [14]

    Scene parsing through ade20k dataset,

    B. Zhou, H. Zhao, X. Puig, S. Fidler, A. Barriuso, and A. Torralba, “Scene parsing through ade20k dataset,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2017, pp. 633– 641

    2017

  7. [15]

    An image is worth 16x16 words: Transformers for image recognition at scale,

    A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly et al., “An image is worth 16x16 words: Transformers for image recognition at scale,” arXiv preprint arXiv:2010.11929 , 2020

    2010 arXiv

  8. [16]

    Interactive graph cuts for optimal bound- ary & region segmentation of objects in nd images,

    Y . Y . Boykov and M.-P. Jolly, “Interactive graph cuts for optimal bound- ary & region segmentation of objects in nd images,” in Proceedings eighth IEEE international conference on computer vision. ICCV 2001 , vol. 1. IEEE, 2001, pp. 105–112

    2001

  9. [17]

    Geodesic star convexity for interactive image segmentation,

    V . Gulshan, C. Rother, A. Criminisi, A. Blake, and A. Zisserman, “Geodesic star convexity for interactive image segmentation,” in 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE, 2010, pp. 3129–3136

    2010

  10. [18]

    Graph cut algorithms for binocular stereo with occlusions,

    V . Kolmogorov and R. Zabih, “Graph cut algorithms for binocular stereo with occlusions,” in Handbook of Mathematical Models in Computer Vision. Springer, 2006, pp. 423–437

    2006

  11. [19]

    Fast approximate energy min- imization via graph cuts,

    Y . Boykov, O. Veksler, and R. Zabih, “Fast approximate energy min- imization via graph cuts,” IEEE Transactions on pattern analysis and machine intelligence, vol. 23, no. 11, pp. 1222–1239, 2001

    2001

  12. [20]

    Geodesic active contours,

    V . Caselles, R. Kimmel, and G. Sapiro, “Geodesic active contours,” International journal of computer vision , vol. 22, pp. 61–79, 1997

    1997

  13. [21]

    Level set methods: an overview and some recent results,

    S. Osher and R. P. Fedkiw, “Level set methods: an overview and some recent results,” Journal of Computational physics , vol. 169, no. 2, pp. 463–502, 2001

    2001

  14. [22]

    Deep interactive object selection,

    N. Xu, B. Price, S. Cohen, J. Yang, and T. S. Huang, “Deep interactive object selection,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 373–381

    2016

  15. [23]

    Pseudoclick: Interactive image segmentation with click imitation,

    Q. Liu, M. Zheng, B. Planche, S. Karanam, T. Chen, M. Niethammer, and Z. Wu, “Pseudoclick: Interactive image segmentation with click imitation,” in European Conference on Computer Vision . Springer, 2022, pp. 728–745

    2022

  16. [24]

    Interactive image segmentation with first click attention,

    Z. Lin, Z. Zhang, L.-Z. Chen, M.-M. Cheng, and S.-P. Lu, “Interactive image segmentation with first click attention,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 13 339–13 348

    2020

  17. [25]

    Conditional diffusion for interactive segmentation,

    X. Chen, Z. Zhao, F. Yu, Y . Zhang, and M. Duan, “Conditional diffusion for interactive segmentation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2021, pp. 7345–7354

    2021

  18. [26]

    Learning spatiotemporal features with 3d convolutional networks,

    D. Tran, L. Bourdev, R. Fergus, L. Torresani, and M. Paluri, “Learning spatiotemporal features with 3d convolutional networks,” in Proceedings of the IEEE international conference on computer vision , 2015, pp. 4489–4497

    2015

  19. [27]

    Quo vadis, action recognition? a new model and the kinetics dataset,

    J. Carreira and A. Zisserman, “Quo vadis, action recognition? a new model and the kinetics dataset,” in proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017, pp. 6299–6308

    2017

  20. [28]

    Slowfast networks for video recognition,

    C. Feichtenhofer, H. Fan, J. Malik, and K. He, “Slowfast networks for video recognition,” in Proceedings of the IEEE/CVF international conference on computer vision , 2019, pp. 6202–6211

    2019

  21. [29]

    Actionclip: A new paradigm for video action recognition,

    M. Wang, J. Xing, and Y . Liu, “Actionclip: A new paradigm for video action recognition,” arXiv preprint arXiv:2109.08472 , 2021

    2021 arXiv

  22. [30]

    Clip4clip: An empirical study of clip for end to end video clip retrieval and captioning,

    H. Luo, L. Ji, M. Zhong, Y . Chen, W. Lei, N. Duan, and T. Li, “Clip4clip: An empirical study of clip for end to end video clip retrieval and captioning,” Neurocomputing, vol. 508, pp. 293–304, 2022

    2022

  23. [31]

    X-clip: End- to-end multi-grained contrastive learning for video-text retrieval,

    Y . Ma, G. Xu, X. Sun, M. Yan, J. Zhang, and R. Ji, “X-clip: End- to-end multi-grained contrastive learning for video-text retrieval,” in Proceedings of the 30th ACM International Conference on Multimedia , 2022, pp. 638–647

    2022

  24. [32]

    ” grabcut

    C. Rother, V . Kolmogorov, and A. Blake, “” grabcut” interactive foreground extraction using iterated graph cuts,” ACM transactions on graphics (TOG), vol. 23, no. 3, pp. 309–314, 2004

    2004

  25. [33]

    A comparative evaluation of interactive segmentation algorithms,

    K. McGuinness and N. E. O’connor, “A comparative evaluation of interactive segmentation algorithms,” Pattern Recognition, vol. 43, no. 2, pp. 434–444, 2010

    2010

  26. [34]

    A benchmark dataset and evaluation methodology for video object segmentation,

    F. Perazzi, J. Pont-Tuset, B. McWilliams, L. Van Gool, M. Gross, and A. Sorkine-Hornung, “A benchmark dataset and evaluation methodology for video object segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 724–732

    2016

  27. [35]

    Microsoft coco: Common objects in context,

    T.-Y . Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Doll ´ar, and C. L. Zitnick, “Microsoft coco: Common objects in context,” in Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13 . Springer,...

    2014

  28. [36]

    Lvis: A dataset for large vocabulary instance segmentation,

    A. Gupta, P. Dollar, and R. Girshick, “Lvis: A dataset for large vocabulary instance segmentation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2019, pp. 5356– 5364

    2019

  29. [37]

    Edgeflow: Achieving practical interactive segmentation with edge-guided flow,

    Y . Hao, Y . Liu, Z. Wu, L. Han, Y . Chen, G. Chen, L. Chu, S. Tang, Z. Yu, Z. Chen et al., “Edgeflow: Achieving practical interactive segmentation with edge-guided flow,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2021, pp. 1551–1560

    2021

  30. [38]

    Efficient mask correction for click-based interactive image segmentation,

    F. Du, J. Yuan, Z. Wang, and F. Wang, “Efficient mask correction for click-based interactive image segmentation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 22 773–22 782

    2023

  31. [39]

    Interformer: Real-time interactive image segmentation,

    Y . Huang, H. Yang, K. Sun, S. Zhang, L. Cao, G. Jiang, and R. Ji, “Interformer: Real-time interactive image segmentation,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 22 301–22 311

    2023

  32. [40]

    Dinov2: Learning robust visual features without supervision,

    M. Oquab, T. Darcet, T. Moutakanni, H. V . V o, M. Szafraniec, V . Khali- dov, P. Fernandez, D. Haziza, F. Massa, A. El-Nouby, R. Howes, P.-Y . Huang, H. Xu, V . Sharma, S.-W. Li, W. Galuba, M. Rabbat, M. Assran, N. Ballas, G. Synnaeve, I. Misra, H. Jegou, J. Mairal, P. Labatu...

    2023

  33. [41]

    Segment everything everywhere all at once,

    X. Zou, J. Yang, H. Zhang, F. Li, L. Li, J. Gao, and Y . J. Lee, “Segment everything everywhere all at once,” arXiv preprint arXiv:2304.06718 , 2023

    2023 arXiv

  34. [42]

    Seggpt: Segmenting everything in context,

    X. Wang, X. Zhang, Y . Cao, W. Wang, C. Shen, and T. Huang, “Seggpt: Segmenting everything in context,” arXiv preprint arXiv:2304.03284 , 2023

    2023 arXiv

  35. [43]

    Segment anything in high quality,

    L. Ke, M. Ye, M. Danelljan, Y . Liu, Y .-W. Tai, C.-K. Tang, and F. Yu, “Segment anything in high quality,” arXiv preprint arXiv:2306.01567 , 2023

    2023 arXiv

  36. [44]

    Image harmonization dataset iharmony4: Hcoco, hadobe5k, hflickr, and hday2night,

    W. Cong, J. Zhang, L. Niu, L. Liu, Z. Ling, W. Li, and L. Zhang, “Image harmonization dataset iharmony4: Hcoco, hadobe5k, hflickr, and hday2night,” arXiv preprint arXiv:1908.10526 , 2019

    1908 arXiv

  37. [45]

    Taming transformers for high- resolution image synthesis,

    P. Esser, R. Rombach, and B. Ommer, “Taming transformers for high- resolution image synthesis,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2021, pp. 12 873–12 883

    2021

  38. [47]

    Modular interactive video object segmentation: Interaction-to-mask, propagation and difference- aware fusion,

    H. K. Cheng, Y .-W. Tai, and C.-K. Tang, “Modular interactive video object segmentation: Interaction-to-mask, propagation and difference- aware fusion,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2021, pp. 5559–5568

    2021

  39. [48]

    Memory aggregation networks for efficient interactive video object segmentation,

    J. Miao, Y . Wei, and Y . Yang, “Memory aggregation networks for efficient interactive video object segmentation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2020, pp. 10 366–10 375

    2020

  40. [49]

    Segmented anisotropic sstem dataset of neural tissue,

    S. Gerhard, J. Funke, J. Martel, A. Cardona, and R. Fetter, “Segmented anisotropic sstem dataset of neural tissue,” figshare, pp. 0–0, 2013

    2013

  41. [50]

    The rsna-asnr-miccai brats 2021 benchmark on brain tumor segmentation and radiogenomic classification,

    U. Baid, S. Ghodasara, S. Mohan, M. Bilello, E. Calabrese, E. Colak, K. Farahani, J. Kalpathy-Cramer, F. C. Kitamura, S. Pati et al. , “The rsna-asnr-miccai brats 2021 benchmark on brain tumor segmentation and radiogenomic classification,” arXiv preprint arXiv:2107.02314 , 2021

    2021 arXiv

  42. [51]

    Automated segmen- tation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the osteoarthritis initia- tive,

    F. Ambellan, A. Tack, M. Ehlke, and S. Zachow, “Automated segmen- tation of knee bone and cartilage combining statistical shape knowledge and convolutional neural networks: Data from the osteoarthritis initia- tive,” Medical image analysis , vol. 52, pp. 109–118, 2019

    2019

Pith tools

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