Pith. sign in

REVIEW 4 major objections 6 minor 50 references

HQ-SMem: Video Segmentation and Tracking Using Memory Efficient Object Embedding With Selective Update and Self-Supervised Distillation Feedback

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read HQ-SMem wraps existing video object segmentation trackers in a SAM-HQ refinement loop gated by DINO-feature similarity, with selective memory pruning and online teacher forcing, to sharpen mask boundaries, resist drift, and keep memory…

desk verdict A genuine plug-in gain for DeAOT-class trackers, with the strongest evidence in the validation ablation — but the memory-efficiency headline rests on an update-rate confound, not on the selective-eviction rule. read the letter →

arxiv 2507.18921 v1 pith:AQPL7ZFR submitted 2025-07-25 cs.CV

Elham Soltani Kazemi , Imad Eddine Toubal , Gani Rahmon , Jaired Collins , K. Palaniappan This is my paper · ORCID
classification cs.CV
keywords videoobjectsegmentationSAM-HQDINOfeaturesimilaritysmartmemoryteacherforcinglong-termtopologicaltransformationDeAOT
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 claims that an existing video object segmentation tracker can be upgraded without retraining it: coarse masks from DeAOT or DMAOT are sent to the high-quality SAM-HQ model, and a DINO-feature similarity check decides whether the refined mask is the same object. The accepted masks are fed back into the tracker's decoder as the next frame's prior, which the authors call online teacher forcing, and a smart memory keeps only non-redundant key frames. The claimed payoff is sharper boundaries, less drift under topological changes and occlusions, and a memory footprint that stays nearly flat instead of growing linearly with video length. On VOTS2024 and VOTSt the method ranks among the top two, and on the Long Video Dataset and LVOS it reports the best region similarity and contour accuracy, about 3 percent above the base DeAOT tracker.

What carries the argument

The load-bearing mechanism is the DINO similarity gate in Eqs. (8)-(9): SAM-HQ yields three proposals per object; the tracker picks the proposal whose DINO feature embedding is closest in cosine similarity to the coarse VOS mask, but accepts it only if the similarity exceeds the threshold $\tau$; otherwise the coarse mask is kept. This gate is what turns SAM's high-resolution edges into a safe improvement rather than a source of object confusion. The companion smart-memory scorer in Eqs. (3)-(7) deletes the stored frame that maximizes relevance times a freshness penalty, and the teacher-forcing update in Eq. (10) makes every accepted refinement the prior for the next frame.

What would settle it

Run HQ-SMem on a sequence with two similar objects that cross or on a topology-split sequence, sweeping the acceptance threshold from 0 to 1, and compare quality and accuracy against the base DeAOT tracker; the paper's claim fails if no threshold gives a consistent win on both split-object and long-video sets. A more direct check is to measure the DINO similarity between true target masks and impostor masks across VOTSt and LVOS: if the two distributions overlap substantially, no threshold can implement the same-object filter the argument assumes.

Watch

Extended reading notes

Core claim

The central discovery is that mask quality and memory efficiency can be improved at inference time by gating a strong segmenter with a self-supervised appearance check. HQ-SMem uses the coarse object masks of DeAOT/DMAOT as prompts to SAM-HQ, producing three mask candidates per object; it then chooses the candidate most similar in DINO feature space to the original VOS mask and replaces the mask only when the cosine similarity is above a threshold. This prevents SAM's prompt ambiguity from pulling the mask onto a neighboring object. The selected mask becomes the decoder's prior for the next frame, so the refinement compounds. In parallel, a selective memory update scores each stored frame by relevance to the current frame and age, removing the frame that is both most similar and oldest, which keeps memory nearly constant. The reported consequence is a consistent gain in quality and accuracy across several DeAOT/DMAOT variants, with a 3 percent improvement in J and F on topology-changing data and new best results on long-video benchmarks.

Load-bearing premise

The load-bearing premise is that a high DINO cosine similarity between a coarse tracker mask and a SAM-HQ proposal always means the proposal is a better mask for the exact same object; if the similarity score cannot separate the tracked object from a look-alike neighbor, or lags behind a sudden split or scale change, the gate admits a wrong mask and the teacher-forcing loop propagates the error.

Editorial extensions

If this is right

  • On topology-change videos, the base tracker's J and F scores rise by about 3 percentage points when wrapped by HQ-SMem.
  • On the Long Video Dataset and LVOS, HQ-SMem reports the best region similarity and contour accuracy among the compared methods, including existing SAM-augmented trackers.
  • The smart memory keeps the end-of-video memory size nearly constant while the base DeAOT's memory grows linearly with frame count, so longer sequences need less extra storage.
  • The full configuration beats both naively passing masks to SAM (which lowers quality) and using only the selective memory, so the DINO gate is necessary for the reported gain.
  • The improvement transfers across the DeAOT and DMAOT family: average quality improves by 4.9 percent and accuracy by 5.6 percent over twelve base configurations.

Reading between the lines

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

  • We infer that the smart-memory rule could likely be grafted onto other memory-based VOS trackers such as XMem or RMem, since it only uses frame embeddings those models already store; the paper does not test that transfer.
  • The acceptance threshold is a single global constant, so a per-object or per-sequence threshold calibrated from the first frame's similarity distribution is a plausible robustness improvement, especially for videos where objects become more alike over time.
  • The paper reports quality and memory but not wall-clock time; the added SAM-HQ and DINO forward passes per frame may dominate runtime, so real-time deployment is an open question.
  • The same DINO-gated refinement idea could be used in prompt-based video editing, where rejecting mask proposals that leave the object's appearance manifold could reduce mask-stitching errors.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper presents HQ-SMem, an inference-time wrapper around DeAOT/DMAOT VOS trackers. Coarse tracker masks are used to prompt HQ-SAM, the resulting SAM proposals are accepted or rejected via DINO feature similarity (Eqs. 8-9), accepted refined masks are fed back into the tracker as an online teacher-forcing prior (Eq. 10), and the tracker memory is managed by a relevance-freshness score with conditional eviction (Eqs. 3-7). Experiments on VOTS2024, VOTSt, Long Video Dataset, and LVOS report consistent Q/Acc gains over the base trackers, a top-two VOTS placement, and an approximately flat memory curve. The paper also includes a component ablation, a prompt ablation, and a limitations section.

Significance. The evaluation uses external held-out benchmarks and pretrained external components, and the multi-backbone ablation in Table 4 shows consistent gains of roughly 2.4-7.6% in Q across DeAOT/DMAOT variants, which is a real strength. The component ablation in the supplement indicates that the HQ-SAM teacher-forcing path is the main quality driver, and the manuscript is transparent about failure cases in the limitations section. If the memory-efficiency and benchmark claims are substantiated, HQ-SMem would be a useful plug-and-play enhancement for existing trackers. However, the memory-efficiency claim is currently confounded by the fixed L/30 update schedule in the evaluation protocol, and the 'new benchmark' claim is made against a dated comparison set.

major comments (4)
  1. [Supplementary C.3, Fig. 4, Supplementary Table 1] The memory-efficiency claim is confounded by the evaluation protocol. Supplementary C.3 states that for VOTS, LVOS, and Long Video Dataset the memory bank update frequency is set to L/30; with updates only once every L/30 frames, even a non-selective uniform-subsampling policy would cap the bank at roughly 30 entries, so the flat blue curve in Fig. 4 does not demonstrate the benefit of the selective scoring function O(z_j) in Eq. (6). The component ablation in Supplementary Table 1 isolates S-Mem at Q 0.636 to 0.639 and Acc 0.751 to 0.758, a gain of only +0.3 Q points, whereas the teacher-forcing component accounts for the larger gains (Q 0.639 to 0.654, Acc 0.758 to 0.788). Please report memory size and quality under a matched update frequency for the base tracker and for a non-selective subsampling baseline, so that the selective-update contribution can be evaluated at fixed update frequency.
  2. [Eq. (7)] The update rule in Eq. (7) does not guarantee the bounded memory that Section 3.2 and Figure 4 claim. When Rel(z_del, z_t) < tau, the rule appends z_t without deleting any frame, so the memory bank can grow without bound if consecutive frames are sufficiently dissimilar. This means the theoretical 'near-constant scaling' statement is not supported by the stated algorithm. Please either prove a bound under the threshold conditions, modify the rule so that a frame is always evicted, or report the actual bank sizes on the long-video benchmarks.
  3. [Tables 2(a)-(b), Abstract] The abstract's claim that HQ-SMem 'sets new benchmarks' on Long Video Dataset and LVOS is not supported by the comparison set in Table 2. The table lists methods up to XMem and RMem and does not include recent published VOS methods evaluated on these benchmarks, such as Cutie, SAM 2, and other 2023-2025 long-video trackers. Without comparisons against the current leaderboard, the 'new benchmark' claim is unsubstantiated; please either add these comparisons or temper the claim to 'improves over the compared baselines.'
  4. [Eqs. (5) and (9), Supplementary Table 2] The method depends on several hand-set hyperparameters, notably the DINO acceptance threshold tau in Eq. (9) and the relevance-freshness trade-off lambda in Eq. (5), but no sensitivity analysis or selection criterion is reported for them. Since these thresholds directly control which SAM proposals are accepted and which frames are evicted, a sensitivity study or at least the chosen values with a justification is needed to establish that the reported gains are robust rather than tuned to the selected benchmarks.
minor comments (6)
  1. [Section 4.5] The heading contains a typo, 'dtate-of-the-art', which should read 'state-of-the-art'.
  2. [Eqs. (8)-(9)] The notation p^c_{i,t} in Eq. (9) is inconsistent with p_{t,i} in Eq. (8), and the superscript c is never defined.
  3. [Section 2.4] The text says experiments are run on DAVIS and YouTubeVOS, but in the experimental setup these datasets are used as training sets rather than evaluation benchmarks; the wording should be revised for accuracy.
  4. [Supplementary C.2] The learning-rate decay schedule is given as a polynomial with power p, but the value of p is not specified.
  5. [Title and Section 3.3] The phrase 'self-supervised distillation feedback' is not an accurate description of the online teacher-forcing mechanism, which feeds HQ-SAM refined masks back into the tracker rather than performing distillation; consider renaming this component.
  6. [Figure 3] The qualitative comparison would be more informative with larger crops and explicit annotations marking the boundary differences described in the text.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: HQ-SMem is evaluated against external held-out benchmarks and its contributions do not reduce to their own inputs by construction.

full rationale

I walked the paper's derivation chain and found no step where a claimed prediction or first-principles result is equivalent to its inputs by construction. The VOS formulation in Eq. (1) and the training objective in Eq. (2) are standard definitions of the external base models, and the S-Mem eviction rule in Eqs. (3)-(7) is a new selection mechanism whose quality is measured against held-out ground truth on VOTS2024, VOTSt, Long Video Dataset, and LVOS. The SAM-HQ proposal generation and DINO verification in Eqs. (8)-(9) are gating rules, not fitted predictions: no parameter is fit to the test set and then reported as a forecast, and the accepted masks are scored externally by IoU, J, and F. The online teacher-forcing loop in Eq. (10) feeds refined masks back into the base tracker, but this is an algorithmic feedback mechanism rather than a logical derivation of the result from itself; whether the feedback helps is determined by independent benchmark measurements. The component ablation in Supplementary Table 1 also isolates S-Mem and SAM-HQTF against baseline DeAOT, so the central quality claims do not collapse into a self-citation chain. The only self-citations, references [27] and [34], appear in related-work context and are not load-bearing for the method's validity. One evaluation concern does exist but it is not circularity: Supplementary Section C.3 sets the memory bank update frequency to L/30 for the long-video benchmarks, so the near-flat memory curve in Figure 4 is partly attributable to the update schedule rather than solely to the selective scoring function O(z_j); this is an attribution or experimental-design issue, not a derivation that reduces to its inputs. Therefore the appropriate circularity score is 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central claim rests on pretrained external components (DeAOT, HQ-SAM, DINO) and on unstated hyperparameters; no new entities are introduced. The most load-bearing premises are the reliability of DINO-based matching for mask acceptance and the assumption that greedy memory removal and teacher-forcing feedback do not accumulate errors.

free parameters (4)
  • lambda (relevance-freshness trade-off in Eq. 5) = not reported
    Controls how strongly older frames are favored for removal; no value or tuning procedure is given, so reimplementation must guess it.
  • tau_mem (memory update threshold in Eq. 7) = not reported
    Determines whether the current frame replaces the most redundant frame; value is never stated.
  • tau_dino (DINO acceptance threshold in Eq. 9) = not reported
    Rejects SAM-HQ proposals whose DINO similarity to the VOS mask is too low; value is never stated.
  • memory update schedules (L/30 for long videos, fixed 5 frames for VOTSt) = L/30 and 5 frames
    Chosen per dataset in Supplementary C.3 and C.5; no sensitivity analysis is provided.
assumptions (4)
  • domain assumption DINO features provide reliable object-level correspondence between coarse VOS masks and SAM-HQ proposals across topological changes and occlusions.
    Used in Eqs. 8-9 to accept or reject SAM proposals; if false, the refinement step propagates wrong masks.
  • domain assumption SAM-HQ prompted with a coarse VOS mask yields a proposal that, when DINO-similar, is more accurate than the coarse mask.
    Underpins the teacher-forcing benefit; the paper's ablation (DeAOT+SAM naive hurts Q) shows this is fragile.
  • domain assumption Greedy removal of the single most redundant frame preserves the information the decoder needs for future frames.
    Eqs. 5-7 assume local greedy selection is near-optimal for memory; no proof or sensitivity analysis.
  • domain assumption Feeding the refined mask y^HQ_{t-1} as the next mask prior does not accumulate errors over long sequences.
    Eq. 10 closes the loop between refinement and propagation; the paper acknowledges mask drift for topological changes in Supplementary F.

how reviews work

0 comments
Cite this review

Pith. "Pith review of HQ-SMem: Video Segmentation and Tracking Using Memory Efficient Object Embedding With Selective Update and Self-Supervised Distillation Feedback." pith.science (2026). https://pith.science/paper/AQPL7ZFR

@misc{pith2026250718921,
  author       = {Pith},
  title        = {Pith review of: HQ-SMem: Video Segmentation and Tracking Using Memory Efficient Object Embedding With Selective Update and Self-Supervised Distillation Feedback},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AQPL7ZFR}},
  note         = {Machine review of arXiv:2507.18921}
}
read the original abstract

Video Object Segmentation (VOS) is foundational to numerous computer vision applications, including surveillance, autonomous driving, robotics and generative video editing. However, existing VOS models often struggle with precise mask delineation, deformable objects, topologically transforming objects, tracking drift and long video sequences. In this paper, we introduce HQ-SMem, for High Quality video segmentation and tracking using Smart Memory, a novel method that enhances the performance of VOS base models by addressing these limitations. Our approach incorporates three key innovations: (i) leveraging SAM with High-Quality masks (SAM-HQ) alongside appearance-based candidate-selection to refine coarse segmentation masks, resulting in improved object boundaries; (ii) implementing a dynamic smart memory mechanism that selectively stores relevant key frames while discarding redundant ones, thereby optimizing memory usage and processing efficiency for long-term videos; and (iii) dynamically updating the appearance model to effectively handle complex topological object variations and reduce drift throughout the video. These contributions mitigate several limitations of existing VOS models including, coarse segmentations that mix-in background pixels, fixed memory update schedules, brittleness to drift and occlusions, and prompt ambiguity issues associated with SAM. Extensive experiments conducted on multiple public datasets and state-of-the-art base trackers demonstrate that our method consistently ranks among the top two on VOTS and VOTSt 2024 datasets. Moreover, HQ-SMem sets new benchmarks on Long Video Dataset and LVOS, showcasing its effectiveness in challenging scenarios characterized by complex multi-object dynamics over extended temporal durations.

Figures

Figures reproduced from arXiv: 2507.18921 by the authors.

Figure 1
Figure 1. Our proposed HQ-SMem selectively stores only key [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. System diagram for our proposed HQ-SMem architecture. The bottom-stream uses DMAOT for multi-object tracking using [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Comparison between DeAOT and our proposed HQ [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Memory usage per frame for DeAOT versus HQ-SMem [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 1
Figure 1. Figure 1: Comparison of DeAOT with our proposed HQ-SMem [PITH_FULL_IMAGE:figures/full_fig_p009_1.png]
Figure 2
Figure 2. Figure 2: Comparison of DeAOT with our proposed HQ-SMem [PITH_FULL_IMAGE:figures/full_fig_p010_2.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 41 canonical work pages

  1. [1]

    XMem++: Production-level video segmenta- tion from few annotated frames

    Maksym Bekuzarov, Ariana Bermudez, Joon-Young Lee, and Hao Li. XMem++: Production-level video segmenta- tion from few annotated frames. In IEEE Int. Conf. Com- puter Vision, pages 635–644, 2023. 2

    work page 2023

  2. [2]

    Object segmentation by long term analysis of point trajectories

    Thomas Brox and Jitendra Malik. Object segmentation by long term analysis of point trajectories. In European Conf. Computer Vision, pages 282–295, 2010. 3

    work page 2010

  3. [3]

    Emerg- ing properties in self-supervised vision transformers

    Mathilde Caron, Hugo Touvron, Ishan Misra, Herv ´e J´egou, Julien Mairal, Piotr Bojanowski, and Armand Joulin. Emerg- ing properties in self-supervised vision transformers. In IEEE Int. Conf. Computer Vision , pages 9650–9660, 2021. 4

    work page 2021

  4. [4]

    XMem: long-term video object segmentation with an Atkinson-Shiffrin mem- ory model

    Ho Kei Cheng and Alexander G Schwing. XMem: long-term video object segmentation with an Atkinson-Shiffrin mem- ory model. In European Conf. Computer Vision, pages 640– 658, 2022. 2, 4, 7

    work page 2022

  5. [5]

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

    Ho Kei Cheng, Yu-Wing Tai, and Chi-Keung Tang. Modular interactive video object segmentation: Interaction-to-mask, propagation and difference-aware fusion. In IEEE Conf. Computer Vision and Pattern Recognition , pages 5559– 5568, 2021. 7

    work page 2021

  6. [6]

    Rethink- ing space-time networks with improved memory coverage for efficient video object segmentation

    Ho Kei Cheng, Yu-Wing Tai, and Chi-Keung Tang. Rethink- ing space-time networks with improved memory coverage for efficient video object segmentation. Advances in Neural Information Processing Systems, 34:11781–11794, 2021. 7

    work page 2021

  7. [7]

    Putting the object back into video object segmentation

    Ho Kei Cheng, Seoung Wug Oh, Brian Price, Joon-Young Lee, and Alexander Schwing. Putting the object back into video object segmentation. In IEEE Conf. Computer Vision and Pattern Recognition, pages 3151–3161, 2024. 2, 8

    work page 2024

  8. [8]

    An image is worth 16x16 words: Trans- formers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, et al. An image is worth 16x16 words: Trans- formers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020. 4

    arXiv 2010

Show all 50 references
  1. [9]

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

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Syl- vain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition ...

    2021

  2. [10]

    Mask r-cnn

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

    2017

  3. [11]

    Lvos: A benchmark for long-term video object segmentation

    Lingyi Hong, Wenchao Chen, Zhongying Liu, Wei Zhang, Pinxue Guo, Zhaoyu Chen, and Wenqiang Zhang. Lvos: A benchmark for long-term video object segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 13480–13492, 2023. 2, 6, 7, 1

    2023

  4. [12]

    Determining opti- cal flow

    Berthold KP Horn and Brian G Schunck. Determining opti- cal flow. Artificial intelligence, 17(1-3):185–203, 1981. 1

    1981

  5. [13]

    Segment anything in high quality

    Lei Ke, Mingqiao Ye, Martin Danelljan, Yu-Wing Tai, Chi- Keung Tang, Fisher Yu, et al. Segment anything in high quality. Advances in Neural Information Processing Systems (NeurIPS), 36, 2024. 3, 5, 8

    2024

  6. [14]

    Segment any- thing

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer White- head, Alexander C Berg, Wan-Yen Lo, et al. Segment any- thing. In IEEE/CVF International Conference on Computer Vision (ICCV), pages 4015–4026, 2023. 3, 8

    2023

  7. [15]

    The first visual object tracking segmentation vots2023 chal- lenge results

    Matej Kristan, Ji ˇr´ı Matas, Martin Danelljan, Michael Fels- berg, Hyung Jin Chang, Luka ˇCehovin Zajc, Alan Luke ˇziˇc, Ondrej Drbohlav, Zhongqun Zhang, Khanh-Tung Tran, et al. The first visual object tracking segmentation vots2023 chal- lenge results. In IEEE/CVF Internatio...

    2023

  8. [16]

    The second visual object tracking seg- mentation vots2024 challenge results, 2024

    Matej Kristan, Jiri Matas, Pavel Tokmakov, Michael Fels- berg, Luka ˇCehovin Zajc, Alan Luke ˇziˇc, Khanh-Tung Tran, Xuan-Son Vu, Johanna Bjorklund, Hyung Jin Chang, and Gustavo Fernandez. The second visual object tracking seg- mentation vots2024 challenge results, 2024. 5, 6,...

    2024

  9. [17]

    Fuxin Li, Taeyoung Kim, Ahmad Humayun, David Tsai, and James M. Rehg. Video segmentation by tracking many figure-ground segments. In 2013 IEEE International Con- ference on Computer Vision, pages 2192–2199, 2013. 3

    2013

  10. [18]

    Video object segmentation with adaptive feature bank and uncertain-region refinement

    Yongqing Liang, Xin Li, Navid Jafari, and Jim Chen. Video object segmentation with adaptive feature bank and uncertain-region refinement. Advances in Neural Informa- tion Processing Systems, 33:3430–3441, 2020. 6, 7, 1, 2

    2020

  11. [19]

    Follow anything: Open- set detection, tracking, and following in real-time

    Alaa Maalouf, Ninad Jadhav, Krishna Murthy Jatavallab- hula, Makram Chahine, Daniel M V ogt, Robert J Wood, An- tonio Torralba, and Daniela Rus. Follow anything: Open- set detection, tracking, and following in real-time. IEEE Robotics and Automation Letters, 9(4):3283–3290, 2024. 4

    2024

  12. [20]

    Background sub- traction in highly dynamic scenes

    Vijay Mahadevan and Nuno Vasconcelos. Background sub- traction in highly dynamic scenes. In 2008 IEEE Confer- ence on Computer Vision and Pattern Recognition, pages 1–

    2008

  13. [21]

    The cell tracking challenge: 10 years of objective benchmarking

    Martin Ma ˇska, Vladim´ır Ulman, Pablo Delgado-Rodriguez, Estibaliz G ´omez-de Mariscal, Tereza Ne ˇcasov´a, Fidel A Guerrero Pe ˜na, Tsang Ing Ren, Elliot M Meyerowitz, Tim Scherr, Katharina L¨offler, et al. The cell tracking challenge: 10 years of objective benchmarking. Nat...

    2023

  14. [22]

    Video object segmentation using space-time memory networks

    Seoung Wug Oh, Joon-Young Lee, Ning Xu, and Seon Joo Kim. Video object segmentation using space-time memory networks. In Proceedings of the IEEE/CVF international conference on computer vision, pages 9226–9235, 2019. 7

    2019

  15. [23]

    Perazzi, J

    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 Computer Vision and Pattern Recognition, 2016. 2, 3, 1

    2016

  16. [25]

    The 2017 davis challenge on video object segmentation.arXiv preprint arXiv:1704.00675, 2017

    Jordi Pont-Tuset, Federico Perazzi, Sergi Caelles, Pablo Ar- bel´aez, Alex Sorkine-Hornung, and Luc Van Gool. The 2017 davis challenge on video object segmentation.arXiv preprint arXiv:1704.00675, 2017. 6

    2017 arXiv

  17. [26]

    Learning transferable visual models from natural language supervi- sion

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, 5 Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervi- sion. In International conference on machine learning...

    2021

  18. [27]

    Deepftsg: Multi-stream asymmetric use-net trellis en- coders with shared decoder feature fusion architecture for video motion segmentation

    Gani Rahmon, Kannappan Palaniappan, Imad Eddine Toubal, Filiz Bunyak, Raghuveer Rao, and Guna Seethara- man. Deepftsg: Multi-stream asymmetric use-net trellis en- coders with shared decoder feature fusion architecture for video motion segmentation. International Journal of Com...

    2024

  19. [28]

    Sam 2: Segment anything in images and videos

    Nikhila Ravi, Valentin Gabeur, Yuan-Ting Hu, Ronghang Hu, Chaitanya Ryali, Tengyu Ma, Haitham Khedr, Roman R¨adle, Chloe Rolland, Laura Gustafson, et al. Sam 2: Segment anything in images and videos. arXiv preprint arXiv:2408.00714, 2024. 1

    2024 arXiv

  20. [29]

    Learning fast and robust target models for video object segmentation

    Andreas Robinson, Felix Jaremo Lawin, Martin Danelljan, Fahad Shahbaz Khan, and Michael Felsberg. Learning fast and robust target models for video object segmentation. In Proceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 7406–7415, 2020. 1

    2020

  21. [30]

    U- net: Convolutional networks for biomedical image segmen- tation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U- net: Convolutional networks for biomedical image segmen- tation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, pa...

    2015

  22. [31]

    A practical adap- tive approach for dynamic background subtraction using an invariant colour model and object tracking

    Bijan Shoushtarian and Helmut E Bez. A practical adap- tive approach for dynamic background subtraction using an invariant colour model and object tracking. Pattern Recog- nition Letters, 26(1):5–26, 2005. 1

    2005

  23. [32]

    Video class agnostic segmentation benchmark for au- tonomous driving

    Mennatullah Siam, Alex Kendall, and Martin Jagersand. Video class agnostic segmentation benchmark for au- tonomous driving. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 2825–2834, 2021. 1

    2021

  24. [33]

    Breaking the” object” in video object segmentation

    Pavel Tokmakov, Jie Li, and Adrien Gaidon. Breaking the” object” in video object segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22836–22845, 2023. 2

    2023

  25. [34]

    Ensemble deep learning object detection fusion for cell tracking, mitosis, and lineage

    Imad Eddine Toubal, Noor Al-Shakarji, DDW Cornelison, and K Palaniappan. Ensemble deep learning object detection fusion for cell tracking, mitosis, and lineage. IEEE Open Journal of Engineering in Medicine and Biology, 2023. 1

    2023

  26. [35]

    Dino-tracker: Taming dino for self-supervised point track- ing in a single video

    Narek Tumanyan, Assaf Singer, Shai Bagon, and Tali Dekel. Dino-tracker: Taming dino for self-supervised point track- ing in a single video. In European Conference on Computer Vision, pages 367–385. Springer, 2025. 4

    2025

  27. [36]

    Attention is all you need

    A Vaswani. Attention is all you need. Advances in Neural Information Processing Systems, 2017. 2

    2017

  28. [37]

    A Comprehensive Review of Modern Object Segmenta- tion Approaches

    Yuanbo Wang, Unaiza Ahsan, Hanyan Li, and Matthew Ha- gen. A Comprehensive Review of Modern Object Segmenta- tion Approaches. 2022. 2

    2022

  29. [38]

    A learning algorithm for continually running fully recurrent neural networks.Neu- ral computation, 1(2):270–280, 1989

    Ronald J Williams and David Zipser. A learning algorithm for continually running fully recurrent neural networks.Neu- ral computation, 1(2):270–280, 1989. 5

    1989

  30. [39]

    Qiangqiang Wu, Tianyu Yang, Wei Wu, and Antoni B. Chan. Scalable video object segmentation with simplified frame- work. In Proceedings of the IEEE/CVF International Con- ference on Computer Vision (ICCV) , pages 13879–13889,

  31. [40]

    Youtube-vos: Sequence-to-sequence video object segmentation

    Ning Xu, Linjie Yang, Yuchen Fan, Jianchao Yang, Dingcheng Yue, Yuchen Liang, Brian Price, Scott Cohen, and Thomas Huang. Youtube-vos: Sequence-to-sequence video object segmentation. In Computer Vision – ECCV , pages 603–619, 2018. 3, 1, 2

    2018

  32. [41]

    Ning Xu, Linjie Yang, Yuchen Fan, Dingcheng Yue, Yuchen Liang, Jianchao Yang, and Thomas S. Huang. Youtube-vos: A large-scale video object segmentation benchmark. CoRR, abs/1809.03327, 2018. 3, 1, 2

    2018 arXiv

  33. [42]

    Learning spatio-temporal transformer for vi- sual tracking

    Bin Yan, Houwen Peng, Jianlong Fu, Dong Wang, and Huchuan Lu. Learning spatio-temporal transformer for vi- sual tracking. In Proceedings of the IEEE/CVF international conference on computer vision, pages 10448–10457, 2021. 8

    2021

  34. [43]

    Decoupling features in hierar- chical propagation for video object segmentation

    Zongxin Yang and Yi Yang. Decoupling features in hierar- chical propagation for video object segmentation. Advances in Neural Information Processing Systems (NeurIPS), 2022. 1, 2, 3, 4, 7, 8

    2022

  35. [44]

    Collaborative video object segmentation by foreground-background inte- gration

    Zongxin Yang, Yunchao Wei, and Yi Yang. Collaborative video object segmentation by foreground-background inte- gration. In European Conference on Computer Vision, pages 332–348. Springer, 2020. 7

    2020

  36. [45]

    Associating ob- jects with transformers for video object segmentation

    Zongxin Yang, Yunchao Wei, and Yi Yang. Associating ob- jects with transformers for video object segmentation. Ad- vances in Neural Information Processing Systems (NeurIPS), 34:2491–2502, 2021. 1, 2, 3, 4, 7, 8

    2021

  37. [46]

    Collabora- tive video object segmentation by multi-scale foreground- background integration

    Zongxin Yang, Yunchao Wei, and Yi Yang. Collabora- tive video object segmentation by multi-scale foreground- background integration. IEEE Transactions on Pattern Anal- ysis and Machine Intelligence, 44(9):4701–4712, 2021. 7

    2021

  38. [47]

    Scalable video object segmen- tation with identification mechanism

    Zongxin Yang, Jiaxu Miao, Yunchao Wei, Wenguan Wang, Xiaohan Wang, and Yi Yang. Scalable video object segmen- tation with identification mechanism. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024. 2

    2024

  39. [48]

    Joint feature learning and relation modeling for tracking: A one-stream framework

    Botao Ye, Hong Chang, Bingpeng Ma, Shiguang Shan, and Xilin Chen. Joint feature learning and relation modeling for tracking: A one-stream framework. In European Conference on Computer Vision, pages 341–357. Springer, 2022. 8

    2022

  40. [49]

    Wide residual net- works

    Sergey Zagoruyko and Nikos Komodakis. Wide residual net- works. arXiv preprint arXiv:1605.07146, 2016. 8

    2016 arXiv

  41. [50]

    RMem: Re- stricted Memory Banks Improve Video Object Segmentation

    Junbao Zhou, Ziqi Pang, and Yu-Xiong Wang. RMem: Re- stricted Memory Banks Improve Video Object Segmentation . In 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 18602–18611, 2024. 2, 7, 1

    2024

  42. [51]

    Tracking anything in high quality, 2023

    Jiawen Zhu, Zhenyu Chen, Zeqi Hao, Shijie Chang, Lu Zhang, Dong Wang, Huchuan Lu, Bin Luo, Jun-Yan He, Jin-Peng Lan, Hanyuan Chen, and Chenyang Li. Tracking anything in high quality, 2023. 3, 4, 8 6

    2023

Pith tools

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