Pith. sign in

REVIEW 3 major objections 2 minor 101 references

RIPPLE turns point tracking annotation into sparse corrections on motion-guided trajectories, matching exhaustive manual quality with 3-25 times fewer clicks.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

RIPPLE recasts annotation as sparse correction on motion-guided trajectories, matching exhaustive manual quality while cutting clicks by 3-25x on five diverse microscopy datasets from jellyfish and sperm.

T0 review reviewed 2026-06-29 challenge →

load-bearing objection RIPPLE gives a workable sparse-correction workflow for hard microscopy tracking cases and reports useful click reductions on real data, but the quality-equivalence claim lacks the metrics needed to evaluate it. the 3 major comments →

arxiv 2605.29220 v1 pith:CSELG3JL submitted 2026-05-28 cs.CV

Motion-guided sparse correction enables expert-quality point tracking across diverse microscopy regimes

Leonidas Zimianitis , Pasindu Thenahandi , Kai Buckhalter , Dineth Jayakody , Julian O. Kimura , Xinyue Liang , Karen Cunningham , Azeem Ahmad
show 5 more authors
Balpreet S. Ahluwalia Sampath Jayarathna Nikos Chrisochoides Brandon Weissbourd Dushan N. Wadduwage
This is my paper
classification cs.CV
keywords point trackingmicroscopy annotationsparse correctionmotion guidancetrajectory estimationbiological imaging
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper presents RIPPLE as a way to annotate point trajectories in difficult microscopy videos without labeling every frame. A user marks only a starting location; the system generates a full path using motion information and the user corrects only the segments that drift. On five datasets covering jellyfish cell movements and fast-moving sperm landmarks, the resulting tracks reached the same accuracy level as complete manual annotation. The approach therefore sits between fully manual labeling and fully automatic tracking, making it feasible to quantify dynamics in videos where exhaustive work is impractical.

Core claim

RIPPLE recasts annotation as sparse correction: a user clicks a starting point, the system proposes a full trajectory via motion guidance, and the user intervenes only where the trajectory drifts. Across the five tested microscopy datasets this produced trajectories equivalent in quality to exhaustive manual annotation while reducing the number of manual clicks by a factor of 3 to 25.

What carries the argument

Motion-guided trajectory proposals that support sparse rather than exhaustive user corrections.

Load-bearing premise

The motion-guided proposals remain accurate enough on the tested regimes that corrections stay sparse instead of requiring near-exhaustive intervention.

What would settle it

A new microscopy sequence in which the proposed trajectories require corrections in nearly every frame would demonstrate that the reported reduction in manual effort does not hold.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Enables immediate quantification of biological dynamics in videos previously too costly to annotate exhaustively.
  • Supports direct benchmarking of automated trackers against the produced trajectories.
  • Generates the gold-standard labeled data required to train and adapt future fully automatic microscopy trackers.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same sparse-correction pattern could be tested on non-microscopy video domains if motion guidance can be adapted.
  • Combining RIPPLE proposals with existing automated trackers might further lower the initial click count.
  • Longer sequences or higher motion complexity would provide a direct test of whether proposal accuracy holds beyond the five evaluated datasets.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 2 minor

Summary. The manuscript introduces RIPPLE (Refinement Interpolation Platform for Point Location Estimation), which reframes point tracking in challenging microscopy videos as sparse user correction of motion-guided trajectory proposals. A user provides an initial click, the system generates a full trajectory, and the user corrects only at drift points. The central claim is that, across five datasets (four from Clytia hemisphaerica and one from sperm landmark tracking), the resulting trajectories match the quality of exhaustive manual annotation while reducing manual clicks by a factor of 3–25×.

Significance. If the equivalence claim holds under rigorous metrics, RIPPLE would supply a scalable intermediate layer between exhaustive manual annotation and fully automatic trackers, directly addressing the annotation bottleneck for non-canonical biological systems. The use of real laboratory datasets from multiple regimes is a concrete strength that supports immediate utility for producing gold-standard data and benchmarking automated methods.

major comments (3)
  1. [Abstract / Results] Abstract and Results section: the claim that RIPPLE 'matched the quality of exhaustive manual annotation' is presented without any stated error metric (endpoint error, tracklet RMSE, or similar), without reporting numerical values against held-out labels, and without statistical tests or inter-annotator agreement baselines. This directly undermines evaluation of the central equivalence result.
  2. [Results] Results section: the reported 3–25× click reduction presupposes that motion-guided proposals remain sufficiently close to ground truth that interventions stay sparse; no quantitative analysis of proposal accuracy (e.g., drift statistics before correction) or cases where corrections become dense is supplied, leaving the sparsity assumption untested.
  3. [Methods / Experiments] Methods / Experiments: no description is given of how the five datasets were split for evaluation, whether any held-out ground truth was used, or how equivalence to exhaustive manual annotation was operationally defined and tested.
minor comments (2)
  1. [Abstract] The range '3 to 25 times' is given without per-dataset breakdown; a table listing clicks per method and dataset would improve transparency.
  2. [Abstract] Dataset descriptions in the abstract are brief; adding one-sentence characterizations of imaging conditions (frame rate, resolution, motion type) would aid readers.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments. We agree that the current manuscript lacks the quantitative details needed to rigorously support the central claims and will revise the text, figures, and methods to address each point.

read point-by-point responses

  1. Referee: [Abstract / Results] Abstract and Results section: the claim that RIPPLE 'matched the quality of exhaustive manual annotation' is presented without any stated error metric (endpoint error, tracklet RMSE, or similar), without reporting numerical values against held-out labels, and without statistical tests or inter-annotator agreement baselines. This directly undermines evaluation of the central equivalence result.

    Authors: We acknowledge that the manuscript presents the equivalence claim without the requested quantitative metrics or statistical comparisons. The original evaluation relied on expert visual assessment across the five laboratory datasets, but this is insufficient for the central result. In revision we will add endpoint error and tracklet RMSE values computed against held-out manual annotations, report numerical results per dataset, include inter-annotator agreement baselines from multiple experts, and apply appropriate statistical tests. revision: yes

  2. Referee: [Results] Results section: the reported 3–25× click reduction presupposes that motion-guided proposals remain sufficiently close to ground truth that interventions stay sparse; no quantitative analysis of proposal accuracy (e.g., drift statistics before correction) or cases where corrections become dense is supplied, leaving the sparsity assumption untested.

    Authors: We agree that the sparsity assumption underlying the click-reduction claim requires explicit quantitative support. The manuscript currently reports only the final click counts without characterizing proposal drift. We will add per-dataset statistics on proposal accuracy (mean and distribution of drift distances before correction), identify any sequences where corrections become dense, and discuss the conditions under which the 3–25× reduction holds. revision: yes

  3. Referee: [Methods / Experiments] Methods / Experiments: no description is given of how the five datasets were split for evaluation, whether any held-out ground truth was used, or how equivalence to exhaustive manual annotation was operationally defined and tested.

    Authors: We will expand the Methods and Experiments sections to specify the evaluation protocol in full. This will include the splitting strategy for the five datasets, confirmation that held-out ground-truth annotations were used for quantitative comparison, and the precise operational definition of equivalence (error thresholds, statistical criteria, and comparison to inter-annotator variability). revision: yes

Circularity Check

0 steps flagged

No circularity; empirical tool description with no derivation chain

full rationale

The paper describes RIPPLE as a sparse-correction annotation platform and reports empirical results on five microscopy datasets, claiming reduced clicks while matching manual annotation quality. No equations, fitted parameters, self-citations, or uniqueness theorems are invoked in any load-bearing way. The central claims rest on direct testing against exhaustive manual labels rather than any reduction to inputs by construction. This is the expected non-finding for a methods/empirical paper without a mathematical derivation.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review yields no identifiable free parameters, axioms, or invented entities; the method is presented as an empirical engineering contribution.

reviewed 2026-06-29 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Motion-guided sparse correction enables expert-quality point tracking across diverse microscopy regimes." pith.science (2026). https://pith.science/paper/CSELG3JL

@misc{pith2026260529220,
  author       = {Pith},
  title        = {Pith review of: Motion-guided sparse correction enables expert-quality point tracking across diverse microscopy regimes},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CSELG3JL}},
  note         = {Machine review of arXiv:2605.29220}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Tracking the dynamics of non-canonical biological systems in microscopy videos remains a persistent challenge. Both classical and learning-based trackers depend on expert-reviewed data to be evaluated and adapted, yet exhaustive manual annotation rarely scales to the videos where these tools are needed most. We developed RIPPLE (Refinement Interpolation Platform for Point Location Estimation), which recasts annotation as sparse correction: a user clicks a starting point, RIPPLE proposes a full trajectory, and the user intervenes only where the trajectory drifts. We tested RIPPLE on five challenging microscopy datasets from our laboratories, four from the transparent jellyfish Clytia hemisphaerica and one tracking landmarks on rapidly moving sperm. Across these, RIPPLE matched the quality of exhaustive manual annotation while reducing manual clicks by 3 to 25 times across datasets. RIPPLE thereby fills a missing layer between manual annotation and fully automated tracking, enabling immediate quantification of biological dynamics, method benchmarking, and the production of the gold-standard data needed to adapt future automated microscopy trackers.

Figures

Figures reproduced from arXiv: 2605.29220 by Azeem Ahmad, Balpreet S. Ahluwalia, Brandon Weissbourd, Dineth Jayakody, Dushan N. Wadduwage, Julian O. Kimura, Kai Buckhalter, Karen Cunningham, Leonidas Zimianitis, Nikos Chrisochoides, Pasindu Thenahandi, Sampath Jayarathna, Xinyue Liang.

Figure 1
Figure 1. Figure 1: RIPPLE workflow, interface, and volumetric track visualization. a, RIPPLE sparse-correction workflow. A user selects a point, RIPPLE estimates a full trajectory automatically, and later corrective insertions trigger local track updating; this process can be repeated until the trajectory is satisfactory. b, Graphical interface of RIPPLE, showing tracking settings, video display, frame navigation, toolbar co… view at source ↗
Figure 2
Figure 2. Figure 2: Dataset diversity and representative sparse-correction behavior. a, Repre￾sentative fields of view from the five microscopy regimes used in this study. b, Example tracked targets at selected time points, illustrating differences in motion, appearance, deformation, crowding, and ambiguity across datasets. c, Representative correction event on a neural trajectory. A user correction inserted at a later frame … view at source ↗
Figure 3
Figure 3. Figure 3: Benchmark summary, sparse-correction scaling, and simple downstream analyses. a, Dataset-level comparison between RIPPLE and exhaustive manual annotation, showing average point precision (APP) versus total manual insertions across the five datasets. b, Neural Clytia dataset comparison of RIPPLE, exhaustive manual annotation, TrackMate, LocoTrack, SLEAP-op, and SLEAP across four practical axes: manual annot… view at source ↗
Figure 4
Figure 4. Figure 4: Annotator disagreement and its contribution to observed APP differ￾ences. a, Representative examples of matched manual insertions from the comparison annotator and the RIPPLE annotator across datasets, illustrating cases of low and high uncertainty about target placement. b, Distribution and cumulative fraction of annotator disagreement measured in pixels. Neural and pinned data show near-subpixel agreemen… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

101 extracted references · 1 canonical work pages

  1. [1]

    Multiscale fluorescence imaging of living samples,

    Y. Wu and H. Shroff, “Multiscale fluorescence imaging of living samples,” Histochem. cell biology158, 301–323 (2022)

    2022

  2. [2]

    Live-cell imaging in the deep learning era,

    J. W. Pylvänäinen, E. Gómez-de Mariscal, R. Henriques, and G. Jacquemet, “Live-cell imaging in the deep learning era,” Curr. Opin. Cell Biol.85, 102271 (2023)

    2023

  3. [3]

    Robust single-particle tracking in live-cell time-lapse sequences,

    K. Jaqaman, D. Loerke, M. Mettlen,et al., “Robust single-particle tracking in live-cell time-lapse sequences,” Nat. methods5, 695–702 (2008)

    2008

  4. [4]

    Objective comparison of particle tracking methods,

    N. Chenouard, I. Smal, F. De Chaumont,et al., “Objective comparison of particle tracking methods,” Nat. methods 11, 281–289 (2014)

    2014

  5. [5]

    Featurepointtrackingandtrajectoryanalysisforvideoimagingincellbiology,

    I.F.SbalzariniandP.Koumoutsakos,“Featurepointtrackingandtrajectoryanalysisforvideoimagingincellbiology,” J. structural biology151, 182–195 (2005)

    2005

  6. [6]

    A review for cell and particle tracking on microscopy images using algorithms and deep learning technologies,

    H.-J. Cheng, C.-H. Hsu, C.-L. Hung, and C.-Y. Lin, “A review for cell and particle tracking on microscopy images using algorithms and deep learning technologies,” biomedical journal45, 465–471 (2022)

    2022

  7. [7]

    A review of progress in single particle tracking: from methods to biophysical insights,

    C. Manzo and M. F. Garcia-Parajo, “A review of progress in single particle tracking: from methods to biophysical insights,” Reports on progress physics78, 124601 (2015)

    2015

  8. [8]

    Live cell imaging analysis of receptor function,

    D. C. Worth and M. Parsons, “Live cell imaging analysis of receptor function,” inLive Cell Imaging: Methods and Protocols,(Springer, 2009), pp. 311–323

    2009

  9. [9]

    Intracellular transport dynamics revealed by single-particle tracking,

    M.-L. Zhang, H.-Y. Ti, P.-Y. Wang, and H. Li, “Intracellular transport dynamics revealed by single-particle tracking,” Biophys. Reports7, 413 (2021)

    2021

  10. [10]

    Phase contrast time-lapse microscopy datasets with automated and manual cell tracking annotations,

    D. F. E. Ker, S. Eom, S. Sanami,et al., “Phase contrast time-lapse microscopy datasets with automated and manual cell tracking annotations,” Sci. data5, 180237 (2018)

    2018

  11. [11]

    Delta: Automated cell segmentation, tracking, and lineage reconstruction using deep learning,

    J.-B. Lugagne, H. Lin, and M. J. Dunlop, “Delta: Automated cell segmentation, tracking, and lineage reconstruction using deep learning,” PLoS computational biology16, e1007673 (2020)

    2020

  12. [12]

    Long-term tracking of budding yeast cells in brightfield microscopy: Cellstar and the evaluation platform,

    C. Versari, S. Stoma, K. Batmanov,et al., “Long-term tracking of budding yeast cells in brightfield microscopy: Cellstar and the evaluation platform,” J. The Royal Soc. Interface14(2017)

    2017

  13. [13]

    Cell population tracking and lineage construction with spatiotemporal context,

    K. Li, E. D. Miller, M. Chen,et al., “Cell population tracking and lineage construction with spatiotemporal context,” Med. image analysis12, 546–566 (2008)

    2008

  14. [14]

    Automated reconstruction of whole-embryo cell lineages by learning from sparse annotations,

    C. Malin-Mayor, P. Hirsch, L. Guignard,et al., “Automated reconstruction of whole-embryo cell lineages by learning from sparse annotations,” Nat. biotechnology41, 44–49 (2023)

    2023

  15. [15]

    Deep neural networks to register and annotate cells in moving and deforming nervous systems,

    A. A. Atanas, A. K.-Y. Lu, B. Goodell,et al., “Deep neural networks to register and annotate cells in moving and deforming nervous systems,” eLife14, RP108159 (2026)

    2026

  16. [16]

    Rethinking deep learning in bioimaging through a data centric lens,

    J. Cao, J. Wenzel, S. Zhang,et al., “Rethinking deep learning in bioimaging through a data centric lens,” npj Imaging 3, 29 (2025)

    2025

  17. [17]

    Trackmate: Anopenandextensibleplatformforsingle-particletracking,

    J.-Y.Tinevez,N.Perry,J.Schindelin,etal.,“Trackmate: Anopenandextensibleplatformforsingle-particletracking,” Methods115, 80–90 (2017)

    2017

  18. [18]

    Trackmate 7: integrating state-of-the-art segmentation algorithms into tracking pipelines,

    D. Ershov, M.-S. Phan, J. W. Pylvänäinen,et al., “Trackmate 7: integrating state-of-the-art segmentation algorithms into tracking pipelines,” Nat. methods19, 829–832 (2022)

    2022

  19. [19]

    u-track3d: Measuring, navigating, and validating dense particle trajectories in three dimensions,

    P. Roudot, W. R. Legant, Q. Zou,et al., “u-track3d: Measuring, navigating, and validating dense particle trajectories in three dimensions,” Cell Reports Methods3(2023)

    2023

  20. [20]

    Organoidtracker: Efficient cell tracking using machine learning and manual error correction,

    R. N. Kok, L. Hebert, G. Huelsz-Prince,et al., “Organoidtracker: Efficient cell tracking using machine learning and manual error correction,” PLoS One15, e0240802 (2020)

    2020

  21. [21]

    An objective comparison of cell-tracking algorithms,

    V. Ulman, M. Maška, K. E. G. Magnusson,et al., “An objective comparison of cell-tracking algorithms,” Nat. methods14, 1141–1152 (2017)

    2017

  22. [22]

    Cellpose: a generalist algorithm for cellular segmentation,

    C. Stringer, T. Wang, M. Michaelos, and M. Pachitariu, “Cellpose: a generalist algorithm for cellular segmentation,” Nat. methods18, 100–106 (2021)

    2021

  23. [23]

    Cell detection with star-convex polygons,

    U. Schmidt, M. Weigert, C. Broaddus, and G. Myers, “Cell detection with star-convex polygons,” inInternational conference on medical image computing and computer-assisted intervention,(Springer, 2018), pp. 265–273

    2018

  24. [24]

    Sleap: A deep learning system for multi-animal pose tracking,

    T. D. Pereira, N. Tabris, A. Matsliah,et al., “Sleap: A deep learning system for multi-animal pose tracking,” Nat. methods19, 486–495 (2022)

    2022

  25. [25]

    Using deeplabcut for 3d markerless pose estimation across species and behaviors,

    T. Nath, A. Mathis, A. C. Chen,et al., “Using deeplabcut for 3d markerless pose estimation across species and behaviors,” Nat. protocols14, 2152–2176 (2019)

    2019

  26. [26]

    TAPIR: Tracking any point with per-frame initialization and temporal refinement,

    C. Doersch, Y. Yang, M. Vecerik,et al., “TAPIR: Tracking any point with per-frame initialization and temporal refinement,” inProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV),(2023), pp. 10061–10072

    2023

  27. [27]

    Cotracker: Itisbettertotracktogether,

    N.Karaev,I.Rocco,B.Graham,etal.,“Cotracker: Itisbettertotracktogether,”inEuropeanconferenceoncomputer vision,(Springer, 2024), pp. 18–35

    2024

  28. [28]

    Local all-pair correspondence for point tracking,

    S. Cho, J. Huang, J. Nam,et al., “Local all-pair correspondence for point tracking,” inEuropean Conference on Computer Vision (ECCV),(2024)

    2024

  29. [29]

    Particle video revisited: Tracking through occlusions using point trajectories,

    A. W. Harley, Z. Fang, and K. Fragkiadaki, “Particle video revisited: Tracking through occlusions using point trajectories,” inECCV,(2022)

    2022

  30. [30]

    Cotracker3: Simpler and better point tracking by pseudo-labelling real videos,

    N. Karaev, Y. Makarov, J. Wang,et al., “Cotracker3: Simpler and better point tracking by pseudo-labelling real videos,” inProceedings of the IEEE/CVF International Conference on Computer Vision,(2025), pp. 6013–6022

    2025

  31. [31]

    Deepprobabilistictrackingofparticlesinfluorescencemicroscopyimages,

    R.Spilger,J.-Y.Lee,V.O.Chagin,etal.,“Deepprobabilistictrackingofparticlesinfluorescencemicroscopyimages,” Med. image analysis72, 102128 (2021)

    2021

  32. [32]

    Robustmodel-basedanalysisofsingle-particletrackingexperiments with spot-on,

    A.S.Hansen, M.Woringer, J.B.Grimm,etal., “Robustmodel-basedanalysisofsingle-particletrackingexperiments with spot-on,” Elife7, e33125 (2018)

    2018

  33. [33]

    Thecelltrackingchallenge: 10yearsofobjectivebenchmarking,

    M.Maška,V.Ulman,P.Delgado-Rodriguez,etal.,“Thecelltrackingchallenge: 10yearsofobjectivebenchmarking,” Nat. Methods20, 1010–1020 (2023)

    2023

  34. [34]

    Pointodyssey: A large-scale synthetic dataset for long-term point tracking,

    Y. Zheng, A. W. Harley, B. Shen,et al., “Pointodyssey: A large-scale synthetic dataset for long-term point tracking,” inProceedings of the IEEE/CVF International Conference on Computer Vision,(2023), pp. 19855–19865

    2023

  35. [35]

    Biological data annotation via a human-augmenting ai-based labeling system,

    D. van der Wal, I. Jhun, I. Laklouk,et al., “Biological data annotation via a human-augmenting ai-based labeling system,” NPJ digital medicine4, 145 (2021)

    2021

  36. [36]

    A benchmark for comparison of cell tracking algorithms,

    M. Maška, V. Ulman, D. Svoboda,et al., “A benchmark for comparison of cell tracking algorithms,” Bioinformatics 30, 1609–1617 (2014)

    2014

  37. [37]

    Tracking cell lineages in 3d by incremental deep learning,

    K. Sugawara, Ç. Çevrim, and M. Averof, “Tracking cell lineages in 3d by incremental deep learning,” Elife11, e69380 (2022)

    2022

  38. [38]

    edetect: afasterrordetectionandcorrectiontoolforlivecellimagingdataanalysis,

    H.Han,G.Wu,Y.Li,andZ.Zi,“edetect: afasterrordetectionandcorrectiontoolforlivecellimagingdataanalysis,” Iscience13, 1–8 (2019)

    2019

  39. [39]

    Tracurate: efficiently curating cell tracks,

    S. Wagner, K. Thierbach, T. Zerjatke,et al., “Tracurate: efficiently curating cell tracks,” SoftwareX13, 100656 (2021)

    2021

  40. [40]

    Segmentation, tracking and cell cycle analysis of live-cell imaging data with cell-acdc,

    F. Padovani, B. Mairhörmann, P. Falter-Braun,et al., “Segmentation, tracking and cell cycle analysis of live-cell imaging data with cell-acdc,” BMC biology20, 174 (2022)

    2022

  41. [41]

    Laptrack: linear assignment particle tracking with tunable metrics,

    Y. T. Fukai and K. Kawaguchi, “Laptrack: linear assignment particle tracking with tunable metrics,” Bioinformatics 39, btac799 (2023)

    2023

  42. [42]

    Automated cell tracking using stardist and trackmate,

    E. Fazeli, N. H. Roy, G. Follain,et al., “Automated cell tracking using stardist and trackmate,” F1000Research9, 1279 (2020)

    2020

  43. [43]

    A genetically tractable jellyfish model for systems and evolutionary neuroscience,

    B. Weissbourd, T. Momose, A. Nair,et al., “A genetically tractable jellyfish model for systems and evolutionary neuroscience,” Cell184, 5854–5868 (2021)

    2021

  44. [44]

    Whole-animal multiplexed single-cell rna-seq reveals transcriptional shifts across clytia medusa cell types,

    T. Chari, B. Weissbourd, J. Gehring,et al., “Whole-animal multiplexed single-cell rna-seq reveals transcriptional shifts across clytia medusa cell types,” Sci. Adv.7, eabh1683 (2021)

    2021

  45. [45]

    Past, present and future of clytia hemisphaerica as a laboratory jellyfish,

    E. Houliston, L. Leclère, C. Munro,et al., “Past, present and future of clytia hemisphaerica as a laboratory jellyfish,” inCurrent topics in developmental biology,vol. 147 (Elsevier, 2022), pp. 121–151

    2022

  46. [46]

    Tap-vid: A benchmark for tracking any point in a video,

    C. Doersch, A. Gupta, L. Markeeva,et al., “Tap-vid: A benchmark for tracking any point in a video,” Adv. Neural Inf. Process. Syst.35, 13610–13626 (2022)

    2022

  47. [48]

    Quantitative phase imaging in biomedicine,

    Y. Park, C. Depeursinge, and G. Popescu, “Quantitative phase imaging in biomedicine,” Nat. photonics12, 578–589 (2018)

    2018

  48. [49]

    Episodic rolling and transient attachments create diversity in sperm swimming behavior,

    D. F. Babcock, P. M. Wandernoth, and G. Wennemuth, “Episodic rolling and transient attachments create diversity in sperm swimming behavior,” BMC biology12, 67 (2014)

    2014

  49. [50]

    Bimodal rheotactic behavior reflects flagellar beat asymmetry in human sperm cells,

    A. Bukatin, I. Kukhtevich, N. Stoop,et al., “Bimodal rheotactic behavior reflects flagellar beat asymmetry in human sperm cells,” Proc. National Acad. Sci.112, 15904–15909 (2015)

    2015

  50. [51]

    Chirality and frequency measurement of longitudinal rolling of human sperm using optical trap,

    Z. Zhong, C. Zhang, R. Liu,et al., “Chirality and frequency measurement of longitudinal rolling of human sperm using optical trap,” Front. Bioeng. Biotechnol.10, 1028857 (2022)

    2022

  51. [52]

    Imagej for microscopy,

    T. J. Collins, “Imagej for microscopy,” Biotechniques43, S25–S30 (2007)

    2007

  52. [54]

    Clytia hemisphaerica: a jellyfish cousin joins the laboratory,

    E. Houliston, T. Momose, and M. Manuel, “Clytia hemisphaerica: a jellyfish cousin joins the laboratory,” Trends Genet.26, 159–167 (2010)

    2010

  53. [55]

    High doses of crispr/cas9 ribonucleoprotein efficiently induce gene knockout with low mosaicism in the hydrozoan clytia hemisphaerica through microhomology-mediated deletion,

    T. Momose, A. De Cian, K. Shiba,et al., “High doses of crispr/cas9 ribonucleoprotein efficiently induce gene knockout with low mosaicism in the hydrozoan clytia hemisphaerica through microhomology-mediated deletion,” Sci. Reports8, 11734 (2018)

    2018

  54. [56]

    The genome of the jellyfish clytia hemisphaerica and the evolution of the cnidarian life-cycle,

    L. Leclère, C. Horin, S. Chevalier,et al., “The genome of the jellyfish clytia hemisphaerica and the evolution of the cnidarian life-cycle,” Nat. ecology & evolution3, 801–810 (2019)

    2019

  55. [57]

    3d slicer: a platform for subject-specific image analysis, visualization, and clinical support,

    R. Kikinis, S. D. Pieper, and K. G. Vosburgh, “3d slicer: a platform for subject-specific image analysis, visualization, and clinical support,” inIntraoperative imaging and image-guided therapy,(Springer, 2013), pp. 277–289

    2013

  56. [58]

    Demo for Annotating a Point Track with Optical Flow,

    Google DeepMind, “Demo for Annotating a Point Track with Optical Flow,” https://github.com/google-deepmind/ tapnet/blob/main/colabs/optical_flow_track_assist.ipynb (2026). Official TAPNet repository. Accessed: 2026-04-26. Motion-guided sparse correction enables expert-quality point tracking across diverse microscopy regimes LEONIDASZIMIANITIS, 1 PASINDUTH...

    2026

  57. [59]

    Fast optical flow using dense inverse search,

    T. Kroeger, R. Timofte, D. Dai, and L. Van Gool, “Fast optical flow using dense inverse search,” inEuropean conference on computer vision,(Springer, 2016), pp. 471–488

    2016

  58. [60]

    Two-frame motion estimation based on polynomial expansion,

    G. Farnebäck, “Two-frame motion estimation based on polynomial expansion,” inScandinavian conference on Image analysis,(Springer, 2003), pp. 363–370

    2003

  59. [61]

    A duality based approach for realtime tv-l 1 optical flow,

    C. Zach, T. Pock, and H. Bischof, “A duality based approach for realtime tv-l 1 optical flow,” inJoint pattern recognition symposium,(Springer, 2007), pp. 214–223

    2007

  60. [62]

    Robust local optical flow for feature tracking,

    T. Senst, V. Eiselein, and T. Sikora, “Robust local optical flow for feature tracking,” IEEE Trans. on Circuits Syst. for Video Technol.22, 1377–1387 (2012)

    2012

  61. [63]

    Deepflow: Large displacement optical flow with deep matching,

    P. Weinzaepfel, J. Revaud, Z. Harchaoui, and C. Schmid, “Deepflow: Large displacement optical flow with deep matching,” inProceedings of the IEEE international conference on computer vision,(2013), pp. 1385–1392

    2013

  62. [64]

    Simpleflow: A non-iterative, sublinear optical flow algorithm,

    M. Tao, J. Bai, P. Kohli, and S. Paris, “Simpleflow: A non-iterative, sublinear optical flow algorithm,” Comput. Graph. Forum31, 345–353 (2012)

    2012

  63. [65]

    Efficient sparse-to-dense optical flow estimation using a learned basis and layers,

    J. Wulff and M. J. Black, “Efficient sparse-to-dense optical flow estimation using a learned basis and layers,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,(2015), pp. 120–130

    2015

  64. [66]

    Bradski, A

    G. Bradski, A. Kaehleret al., “Opencv,” Dr. Dobb’s journal software tools3, 1–81 (2000)

    2000

  65. [67]

    Denseopticalflowbyiterativelocalwindowregistration,

    G.LeBesneraisandF.Champagnat,“Denseopticalflowbyiterativelocalwindowregistration,”inIEEEInternational Conference on Image Processing 2005,vol. 1 (IEEE, 2005), pp. I–137

    2005

  66. [68]

    Determining optical flow,

    B. K. Horn and B. G. Schunck, “Determining optical flow,” Artif. intelligence17, 185–203 (1981)

    1981

  67. [69]

    RAFT: Recurrent all-pairs field transforms for optical flow,

    Z. Teed and J. Deng, “RAFT: Recurrent all-pairs field transforms for optical flow,” inEuropean Conference on Computer Vision (ECCV),(Springer, 2020), pp. 402–419

    2020

  68. [70]

    Learning to estimate hidden motions with global motion aggregation,

    S. Jiang, D. Campbell, Y. Lu,et al., “Learning to estimate hidden motions with global motion aggregation,” in Proceedings of the IEEE/CVF international conference on computer vision,(2021), pp. 9772–9781

    2021

  69. [71]

    Flowformer: Atransformerarchitectureforopticalflow,

    Z.Huang,X.Shi,C.Zhang,etal.,“Flowformer: Atransformerarchitectureforopticalflow,”inEuropeanconference on computer vision,(Springer, 2022), pp. 668–685

    2022

  70. [72]

    Flowformer++: Masked cost volume autoencoding for pretraining optical flow estimation,

    X. Shi, Z. Huang, D. Li,et al., “Flowformer++: Masked cost volume autoencoding for pretraining optical flow estimation,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition,(2023), pp. 1599–1610

    2023

  71. [73]

    Gmflow: Learningopticalflowviaglobalmatching,

    H.Xu,J.Zhang,J.Cai,etal.,“Gmflow: Learningopticalflowviaglobalmatching,”inProceedingsoftheIEEE/CVF conference on computer vision and pattern recognition,(2022), pp. 8121–8130

    2022

  72. [74]

    Unifying flow, stereo and depth estimation,

    H. Xu, J. Zhang, J. Cai,et al., “Unifying flow, stereo and depth estimation,” IEEE Trans. on Pattern Anal. Mach. Intell.45, 13941–13958 (2023)

    2023

  73. [75]

    Sea-raft: Simple, efficient, accurate raft for optical flow,

    Y. Wang, L. Lipson, and J. Deng, “Sea-raft: Simple, efficient, accurate raft for optical flow,” inEuropean Conference on Computer Vision,(Springer, 2024), pp. 36–54

    2024

  74. [76]

    High resolution multi-scale raft (robust vision challenge 2022),

    A. Jahedi, M. Luz, L. Mehl,et al., “High resolution multi-scale raft (robust vision challenge 2022),” arXiv preprint arXiv:2210.16900 (2022)

    work page arXiv 2022

  75. [77]

    Neuflow-v2: Push high-efficiency optical flow to the limit,

    Z. Zhang, A. Gupta, H. Jiang, and H. Singh, “Neuflow-v2: Push high-efficiency optical flow to the limit,” in2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS),(IEEE, 2025), pp. 2479–2485

    2025

  76. [78]

    Ccmr: High resolution optical flow estimation via coarse-to-fine context-guided motion reasoning,

    A. Jahedi, M. Luz, M. Rivinius, and A. Bruhn, “Ccmr: High resolution optical flow estimation via coarse-to-fine context-guided motion reasoning,” inProceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision,(2024), pp. 6899–6908

    2024

  77. [79]

    Craft: Cross-attentional flow transformer for robust optical flow,

    X. Sui, S. Li, X. Geng,et al., “Craft: Cross-attentional flow transformer for robust optical flow,” inProceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition,(2022), pp. 17602–17611

    2022

  78. [80]

    Memflow: Optical flow estimation and prediction with memory,

    Q. Dong and Y. Fu, “Memflow: Optical flow estimation and prediction with memory,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition,(2024), pp. 19068–19078

    2024

  79. [81]

    Skflow: Learning optical flow with super kernels,

    S. Sun, Y. Chen, Y. Zhu,et al., “Skflow: Learning optical flow with super kernels,” Adv. Neural Inf. Process. Syst. 35, 11313–11326 (2022)

    2022

  80. [82]

    Recurrent partial kernel network for efficient optical flow estimation,

    H. Morimitsu, X. Zhu, X. Ji, and X.-C. Yin, “Recurrent partial kernel network for efficient optical flow estimation,” inProceedings of the AAAI Conference on Artificial Intelligence,vol. 38 (2024), pp. 4278–4286

    2024

Showing first 80 references.

This paper was first reviewed by grok-4.3 on June 29, 2026.