Pith. sign in

REVIEW 4 major objections 6 minor 79 references

Triangular Consistency as a Universal Constraint for Learning Optical Flow

T0 review · 4 major / 6 minor · reviewed 2026-07-12 · grok-4.5

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

Pith's one-line read Optical flow fields must compose; enforcing that triangle of agreement is a free, architecture-agnostic training signal that improves accuracy and transfer.

desk verdict Clean geometric plug-in that unifies cycle/temporal/asymmetric-aug under one residual and shows real multi-regime gains; soft spots are single-run tables and the single-layer assumption, not the core idea. read the letter →

arxiv 2606.19938 v3 pith:KMTTZQHY submitted 2026-06-18 cs.CV cs.AI

Yi Xiao , Carlos Rodriguez Coronel , Jing Zhan , Haniyeh Ehsani Oskouie , Alex Wong , Dong Lao This is my paper · ORCID
classification cs.CVcs.AI
keywords opticalflowtriangularconsistencycompositionalconstraintself-supervisiondataaugmentationcycletemporalchainingdomainadaptation
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

Optical flow is a continuous non-rigid mapping of scene coordinates. Any two successive mappings must compose to the direct mapping between the first and last frame. The paper turns this elementary geometric fact into a training loss called triangular consistency: compose two estimated flows, compare the result with the third, and penalize the residual. The same rule recovers ordinary cycle consistency, longer-range temporal chaining, and analytic data augmentation under controlled synthetic transforms. Because the constraint is purely geometric, it needs no extra labels, almost no extra compute, and works with any network or supervision regime. Experiments show consistent gains when it is simply added to existing pipelines for supervised, unsupervised, and few-shot adaptation training.

What carries the argument

Triangular consistency loss: form the residual between a directly estimated flow and the composition of two intermediate flows, mask unreliable (occluded) pixels via forward-backward checks, and penalize the residual with a robust norm. Instantiations cover cycle consistency, temporal chaining, and analytic affine augmentation.

What would settle it

Add the same triangular losses (with the paper’s occlusion mask and weights) to a standard unsupervised or supervised optical-flow trainer on Sintel or KITTI; if endpoint error and cross-dataset transfer do not improve relative to the identical baseline without those losses, the claimed utility of the constraint is false.

Watch

Extended reading notes

Core claim

The authors show that the simplest non-trivial compositional relation among three optical-flow fields—triangular consistency—supplies a universal, first-principled supervision signal. When two flows are composed to induce a third, enforcing agreement among the three improves accuracy and cross-dataset generalization across supervised, unsupervised, and transfer settings without changing the estimator or requiring new annotations.

Load-bearing premise

After simple forward-backward occlusion masking, a single-layer correspondence still holds on enough pixels for the residual to be a useful training signal rather than noise or bias.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes triangular consistency: compose two optical-flow fields to induce a third and penalize the residual among the three (Eqs. 1–3). The same geometric rule is instantiated as (i) forward–backward cycle consistency, (ii) temporal chaining on frame triplets, and (iii) analytic pseudo-labels under asymmetric affine augmentation of only the target frame. The resulting losses are architecture-agnostic, label-free, and claimed to add negligible overhead. Empirically, the authors plug the losses into ARFlow (unsupervised) and RAFT (self-supervised adaptation and supervised training) and report gains on Sintel, KITTI, HD1K, and Middlebury, including up to 18.1% EPE reduction under single-epoch unlabeled adaptation, 6–8% under unsupervised training, and up to 23.1% cross-dataset transfer under supervised training. Ablations (Tab. 3) and a limitations discussion (Sec. 5.3) accompany the main tables.

Significance. If the reported gains hold under multi-seed evaluation, the work supplies a simple, first-principled, plug-and-play training signal that is orthogonal to architecture and to photometric losses. The analytic asymmetric-augmentation construction (Sec. 3.3, Eqs. 5–8) is particularly useful: it produces exact pseudo-ground-truth without resampling artifacts and expands motion statistics without a simulator. Code is released. The contribution is incremental rather than foundational—composition and cycle consistency are classical—but the systematic treatment across three regimes and the closed-form augmentation are practically valuable for optical-flow training and domain adaptation.

major comments (4)
  1. Tables 1–4 report single-run point estimates with no error bars, multi-seed averages, or statistical tests. The headline adaptation result (Tab. 1: 18.1% / 15.4% after 45 iterations) is especially sensitive to seed and batch composition; without variance it is hard to judge whether the gain is reliable or a lucky trajectory. Multi-seed means and standard deviations (or at least three independent runs) for the main tables are needed to support the “consistent improvement” claim.
  2. Sec. 4.1 adapts on Sintel’s unlabeled test split and evaluates on the labeled training split, with batch-norm statistics frozen. This is an unconventional train/eval swap justified only by server constraints. The protocol should be stated more prominently as a controlled diagnostic rather than a standard test-time adaptation benchmark, and at least one conventional split (or a held-out subset of the training split) should be reported so readers can compare against prior adaptation work.
  3. Sec. 3.1 and Limitations 5.3 acknowledge that single-layer composition fails under occlusion and multi-layer motion, and that M is built from forward–backward checks. The paper never quantifies the fraction of pixels retained by M, nor the residual error of composition on the masked support, on Sintel vs. KITTI. Without that measurement it is difficult to assess how often L_tri (Eq. 3) is true geometric supervision versus soft noise—especially for the “universal” claim when multi-layer motion is common.
  4. The abstract and introduction call the method a “universal” plug-and-play component, yet experiments use only ARFlow and RAFT. Both already incorporate related consistency ideas; transfer to a modern transformer-style estimator (e.g., FlowFormer / SEA-RAFT) or to a pure supervised baseline without photometric terms would better support architecture-agnostic generality. Softening “universal” to “architecture-agnostic within the tested family” or adding one additional backbone would align the claim with the evidence.
minor comments (6)
  1. Fig. 1 caption and body use “T riangular” with a stray space; fix throughout.
  2. Eq. (3) uses ρ(·) without specifying the concrete robust norm used in experiments (Charbonnier, Huber, L1?); state it explicitly in Sec. 3.3.
  3. Affine sampling ranges (translation, rotation, scale) for L_aug are free parameters but never listed; add them to the implementation or appendix.
  4. Tab. 3: the row “+ Aug + Temp + Cyc (λ_aug=0.02, …)” is worse than λ_aug=0.01; a one-sentence note on weight sensitivity would help practitioners.
  5. Related work could more sharply contrast Jeong et al. (CVPR 2022) “Imposing consistency” and SMURF’s multi-frame terms against the asymmetric analytic augmentation claimed as novel here.
  6. Sec. 3.3 speed numbers (0.00030 s, 0.12% wall-clock) are hardware-specific; reporting relative FLOPs or a second GPU would make the “negligible overhead” claim more portable.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: triangular consistency is a geometric definition used as an empirical training loss; reported gains are ordinary held-out EPE, not forced by construction or self-citation.

full rationale

The paper defines triangular residual r = v_{t,t+2} - (v_{t,t+1} + v_{t+1,t+2}(x + v_{t,t+1})) directly from the composition of displacement fields (Eqs. 1-2) and turns it into a robust loss (Eq. 3) with an occlusion mask. Cycle, temporal, and analytic-augmentation variants are special cases of the same identity. This is a constraint, not a prediction derived from fitted parameters. All experimental claims are ordinary endpoint-error or Fl-all numbers on held-out splits (Sintel, KITTI, HD1K, Middlebury) under supervised, unsupervised, and single-epoch adaptation regimes; no constant is fitted on a subset and then re-presented as a forecast, and no uniqueness theorem or load-bearing self-citation is required for the geometric identity. Self-citations appear only in related-work and acknowledgements for prior tracking/composition work and do not underwrite the loss or the numbers. The derivation chain is therefore self-contained and non-circular.

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

The central claim rests on one geometric domain assumption (composition of displacement fields) plus standard occlusion heuristics and a handful of ordinary loss-weight free parameters. No new physical entities are postulated; everything else is either standard mathematics or empirical tuning.

free parameters (3)
  • λ_aug, λ_temp, λ_cyc = 0.01 / 0.003 / 0.005 (best reported)
    Relative weights of the three consistency terms; chosen by grid search on Sintel (Tab. 3) and affect final accuracy.
  • affine sampling ranges (translation, rotation, scale)
    Control the diversity of synthetic motions used for augmentation; exact ranges not stated in the text yet influence generalization claims.
  • EMA decay for teacher network
    Stabilizes self-supervised adaptation; value left unspecified.
assumptions (3)
  • domain assumption Displacement fields compose: v_{t,t+2}(x) ≈ v_{t,t+1}(x) + v_{t+1,t+2}(x + v_{t,t+1}(x)) in co-visible regions
    Stated as first-principled geometry in Sec. 1 and formalized in Eqs. 1–2; holds only where a single-layer correspondence exists.
  • domain assumption Forward-backward residual sufficiently identifies occluded or invalid pixels for soft masking
    Sec. 3.1 and 3.3; standard heuristic, not proven complete.
  • standard math An affine map induces an exact closed-form flow that can be used as ground truth without resampling artifacts
    Derived in Sec. 3.3 Eqs. 5–8; mathematically correct for affine transforms.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Triangular Consistency as a Universal Constraint for Learning Optical Flow." pith.science (2026). https://pith.science/paper/KMTTZQHY

@misc{pith2026260619938,
  author       = {Pith},
  title        = {Pith review of: Triangular Consistency as a Universal Constraint for Learning Optical Flow},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KMTTZQHY}},
  note         = {Machine review of arXiv:2606.19938}
}
read the original abstract

We propose triangular consistency as a first-principled constraint for optical flow, which is agnostic to network architecture, supervision type, and dataset, and applies to both image-pair and multi-frame settings. This simple but powerful constraint is to compose two flows to induce a third flow and enforce consistency among the three. The composed flows may arise from (i) image pairs, yielding cycle consistency; (ii) multiple video frames, producing longer-range motion through temporal chaining; or (iii) image pairs combined with controlled synthetic transformations, which becomes data augmentation. This triangular consistency introduces negligible computational overhead and requires no additional annotations. Since it is derived directly from the geometry of optical flow, it does not rely on model-specific assumptions and serves as a ``universal'' plug-and-play component for optical flow training. Experiments show consistent improvement across supervised, unsupervised, and transfer learning settings.

Figures

Figures reproduced from arXiv: 2606.19938 by the authors.

Figure 1
Figure 1. Triangular consistency as a universal compositional principle. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Self-supervised adaptation with triangular consistency. [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Qualitative comparison on MPI-Sintel. Regions highlighted by dashed cir￾cles illustrate typical improvements introduced by triangular consistency. Our method produces motion fields that better align with the structural boundaries and motion continuity of the scene, reflecting the geometric consistency enforced during training. 4.2 Unsupervised Training Encouraged by the improvements observed in the self-supervised a… view at source ↗

Review history (3 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

79 extracted references · 1 linked inside Pith

  1. [1]

    In: Proceedings of the IEEE international conference on computer vision

    Bailer, C., Taetz, B., Stricker, D.: Flow fields: Dense correspondence fields for highly accurate large displacement optical flow estimation. In: Proceedings of the IEEE international conference on computer vision. pp. 4015–4023 (2015)

    2015

  2. [2]

    International journal of computer vision92(1), 1–31 (2011)

    Baker, S., Scharstein, D., Lewis, J.P., Roth, S., Black, M.J., Szeliski, R.: A database and evaluation methodology for optical flow. International journal of computer vision92(1), 1–31 (2011)

    2011

  3. [3]

    IEEE transactions on pattern analysis and machine intelligence33(3), 500–513 (2010)

    Brox, T., Malik, J.: Large displacement optical flow: descriptor matching in vari- ational motion estimation. IEEE transactions on pattern analysis and machine intelligence33(3), 500–513 (2010)

    2010

  4. [4]

    In: European conference on computer vision

    Butler, D.J., Wulff, J., Stanley, G.B., Black, M.J.: A naturalistic open source movie for optical flow evaluation. In: European conference on computer vision. pp. 611–

  5. [5]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Chen, Q., Koltun, V.: Full flow: Optical flow estimation by global optimization over regular grids. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 4706–4714 (2016)

    2016

  6. [6]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Chen, Z., Jin, H., Lin, Z., Cohen, S., Wu, Y.: Large displacement optical flow from nearest neighbor fields. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 2443–2450 (2013)

    2013

  7. [7]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Cho, S., Huang, J., Kim, S., Lee, J.Y.: Flowtrack: Revisiting optical flow for long- range dense tracking. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 19268–19277 (2024)

    2024

  8. [8]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Doersch, C., Yang, Y., Vecerik, M., Gokay, D., Gupta, A., Aytar, Y., Carreira, J., Zisserman, A.: Tapir: Tracking any point with per-frame initialization and tem- poral refinement. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 10061–10072 (2023)

    2023

Show all 79 references
  1. [9]

    In: Proceedings of the IEEE/CVF Conference on computer vision and pattern recognition

    Dong, Q., Cao, C., Fu, Y.: Rethinking optical flow from geometric matching consis- tent perspective. In: Proceedings of the IEEE/CVF Conference on computer vision and pattern recognition. pp. 1337–1347 (2023)

    2023

  2. [10]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

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

    2024

  3. [11]

    In: Proceedings of the IEEE international conference on computer vision

    Dosovitskiy, A., Fischer, P., Ilg, E., Hausser, P., Hazirbas, C., Golkov, V., Van Der Smagt, P., Cremers, D., Brox, T.: Flownet: Learning optical flow with convolu- tional networks. In: Proceedings of the IEEE international conference on computer vision. pp. 2758–2766 (2015)

    2015

  4. [12]

    IEEE Robotics and Automation Letters4(2), 1661–1668 (2019)

    Fei, X., Wong, A., Soatto, S.: Geo-supervised visual depth prediction. IEEE Robotics and Automation Letters4(2), 1661–1668 (2019)

    2019

  5. [13]

    In: Asian Conference on Computer Vision

    Garg, R., Pizarro, L., Rueckert, D., Agapito, L.: Dense multi-frame optic flow for non-rigid objects using subspace constraints. In: Asian Conference on Computer Vision. pp. 460–473. Springer (2010)

    2010

  6. [14]

    In: 2012 IEEE conference on computer vision and pattern recognition

    Geiger, A., Lenz, P., Urtasun, R.: Are we ready for autonomous driving? the kitti vision benchmark suite. In: 2012 IEEE conference on computer vision and pattern recognition. pp. 3354–3361. IEEE (2012)

    2012

  7. [15]

    Gibson, J.J.: The perception of the visual world. (1950)

    1950

  8. [16]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    He, K., Fan, H., Wu, Y., Xie, S., Girshick, R.: Momentum contrast for unsupervised visual representation learning. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 9729–9738 (2020) Triang. Consist. as Univ. Constraint for Learn. Optica...

    2020

  9. [17]

    Horn,B.K.,Schunck,B.G.:Determiningopticalflow.Artificialintelligence17(1-3), 185–203 (1981)

    1981

  10. [18]

    In: European conference on computer vision

    Huang, Z., Shi, X., Zhang, C., Wang, Q., Cheung, K.C., Qin, H., Dai, J., Li, H.: Flowformer: A transformer architecture for optical flow. In: European conference on computer vision. pp. 668–685. Springer (2022)

    2022

  11. [19]

    In: Pro- ceedings of the Seventh IEEE International Conference on Computer Vision

    Irani, M.: Multi-frame optical flow estimation using subspace constraints. In: Pro- ceedings of the Seventh IEEE International Conference on Computer Vision. vol. 1, pp. 626–633. IEEE (1999)

    1999

  12. [20]

    International Journal of Computer Vision79(1), 71–84 (2008)

    Jackson, J.D., Yezzi, A.J., Soatto, S.: Dynamic shape and appearance modeling via moving and deforming layers. International Journal of Computer Vision79(1), 71–84 (2008)

    2008

  13. [21]

    In: Proceedings of the European confer- ence on computer vision (ECCV)

    Janai, J., Guney, F., Ranjan, A., Black, M., Geiger, A.: Unsupervised learning of multi-frame optical flow with occlusions. In: Proceedings of the European confer- ence on computer vision (ECCV). pp. 690–706 (2018)

    2018

  14. [22]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Janai, J., Guney, F., Wulff, J., Black, M.J., Geiger, A.: Slow flow: Exploiting high- speed cameras for accurate and diverse optical flow reference data. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 3597– 3607 (2017)

    2017

  15. [23]

    In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition

    Jeong, J., Lin, J.M., Porikli, F., Kwak, N.: Imposing consistency for optical flow estimation. In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition. pp. 3181–3191 (2022)

    2022

  16. [24]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition

    Jiang, S., Lu, Y., Li, H., Hartley, R.: Learning optical flow from a few matches. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition. pp. 16592–16600 (2021)

    2021

  17. [25]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Jiao, Y., Tran, T.D., Shi, G.: Effiscene: Efficient per-pixel rigidity inference for unsupervised joint learning of optical flow, depth, camera pose and motion seg- mentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 5538–5547 (2021)

    2021

  18. [26]

    In: European conference on computer vision

    Jonschkowski, R., Stone, A., Barron, J.T., Gordon, A., Konolige, K., Angelova, A.: What matters in unsupervised optical flow. In: European conference on computer vision. pp. 557–572. Springer (2020)

    2020

  19. [27]

    In: European conference on computer vision

    Karaev, N., Rocco, I., Graham, B., Neverova, N., Vedaldi, A., Rupprecht, C.: Cotracker: It is better to track together. In: European conference on computer vision. pp. 18–35. Springer (2024)

    2024

  20. [28]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops

    Kondermann,D.,Nair,R.,Honauer,K.,Krispin,K.,Andrulis,J.,Brock,A.,Gusse- feld, B., Rahimimoghaddam, M., Hofmann, S., Brenner, C., et al.: The hci bench- mark suite: Stereo and flow ground truth with uncertainties for urban autonomous driving. In: Proceedings of the IEEE Confere...

    2016

  21. [29]

    In: Pro- ceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Lao, D., Sundaramoorthi, G.: Minimum delay moving object detection. In: Pro- ceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 4250–4259 (2017)

    2017

  22. [30]

    In: Pro- ceedings of the European conference on computer vision (ECCV)

    Lao, D., Sundaramoorthi, G.: Extending layered models to 3d motion. In: Pro- ceedings of the European conference on computer vision (ECCV). pp. 435–451 (2018)

    2018

  23. [31]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Lao, D., Wang, C., Wong, A., Soatto, S.: Diffeomorphic template registration for atmospheric turbulence mitigation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 25107–25116 (2024)

    2024

  24. [32]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Lao, D., Zhu, P., Wonka, P., Sundaramoorthi, G.: Flow-guided video inpainting with scene templates. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 14599–14608 (2021) 18 Y. Xiao et al

    2021

  25. [33]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Le Moing, G., Ponce, J., Schmid, C.: Dense optical tracking: Connecting the dots. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 19187–19197 (2024)

    2024

  26. [34]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Liu, L., Zhang, J., He, R., Liu, Y., Wang, Y., Tai, Y., Luo, D., Wang, C., Li, J., Huang, F.: Learning by analogy: Reliable supervision from transformations for unsupervised optical flow estimation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern reco...

    2020

  27. [35]

    In: Proceedings of the AAAI conference on artificial intelligence

    Liu, P., King, I., Lyu, M.R., Xu, J.: Ddflow: Learning optical flow with unlabeled data distillation. In: Proceedings of the AAAI conference on artificial intelligence. vol. 33, pp. 8770–8777 (2019)

    2019

  28. [36]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (June 2019)

    Liu, P., Lyu, M., King, I., Xu, J.: Selflow: Self-supervised learning of optical flow. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (June 2019)

    2019

  29. [37]

    In: Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part II

    Liu, T.Y., Agrawal, P., Chen, A., Hong, B.W., Wong, A.: Monitored distillation for positive congruent depth completion. In: Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part II. pp. 35–53. Springer (2022)

    2022

  30. [38]

    In: IJCAI’81: 7th international joint conference on Artificial intelligence

    Lucas, B.D., Kanade, T.: An iterative image registration technique with an appli- cation to stereo vision. In: IJCAI’81: 7th international joint conference on Artificial intelligence. vol. 2, pp. 674–679 (1981)

    1981

  31. [39]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Mayer, N., Ilg, E., Hausser, P., Fischer, P., Cremers, D., Dosovitskiy, A., Brox, T.: A large dataset to train convolutional networks for disparity, optical flow, and scene flow estimation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. ...

    2016

  32. [40]

    In: Proceedings of the AAAI conference on artificial intelligence

    Meister, S., Hur, J., Roth, S.: Unflow: Unsupervised learning of optical flow with a bidirectional census loss. In: Proceedings of the AAAI conference on artificial intelligence. vol. 32 (2018)

    2018

  33. [41]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Menze, M., Geiger, A.: Object scene flow for autonomous vehicles. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 3061–3070 (2015)

    2015

  34. [42]

    In: Pro- ceedingsoftheIEEE/CVFWinterConferenceonApplicationsofComputerVision

    Neoral, M., Šer` ych, J., Matas, J.: Mft: Long-term tracking of every pixel. In: Pro- ceedingsoftheIEEE/CVFWinterConferenceonApplicationsofComputerVision. pp. 6837–6847 (2024)

    2024

  35. [43]

    The Fourteenth International Conference on Learning Representations (2026)

    Park,H.,Chen,R.,Rim,P.,Lao,D.,Wong,A.:Orcas:Unsuperviseddepthcomple- tion via occluded region completion as supervision. The Fourteenth International Conference on Learning Representations (2026)

    2026

  36. [44]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Poggi, M., Tosi, F.: Flowseek: optical flow made easier with depth foundation mod- els and motion bases. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 5667–5679 (2025)

    2025

  37. [45]

    In: Proceedings of the AAAI conference on artificial intelligence

    Ren, Z., Yan, J., Ni, B., Liu, B., Yang, X., Zha, H.: Unsupervised deep learning for optical flow estimation. In: Proceedings of the AAAI conference on artificial intelligence. vol. 31 (2017)

    2017

  38. [46]

    In: Proceedings of the IEEE con- ference on computer vision and pattern recognition

    Revaud, J., Weinzaepfel, P., Harchaoui, Z., Schmid, C.: Epicflow: Edge-preserving interpolation of correspondences for optical flow. In: Proceedings of the IEEE con- ference on computer vision and pattern recognition. pp. 1164–1172 (2015)

    2015

  39. [47]

    In: 2012 IEEE Conference on Computer Vision and Pattern Recognition

    Ricco, S., Tomasi, C.: Dense lagrangian motion estimation with occlusions. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition. pp. 1800–1807. IEEE (2012)

    2012

  40. [48]

    International journal of computer vision80(1), 72–91 (2008) Triang

    Sand, P., Teller, S.: Particle video: Long-range motion estimation using point tra- jectories. International journal of computer vision80(1), 72–91 (2008) Triang. Consist. as Univ. Constraint for Learn. Optical Flow 19

    2008

  41. [49]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Shi, X., Huang, Z., Bian, W., Li, D., Zhang, M., Cheung, K.C., See, S., Qin, H., Dai, J., Li, H.: Videoflow: Exploiting temporal cues for multi-frame optical flow estimation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 12469–12480 (2023)

    2023

  42. [50]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Stone, A., Maurer, D., Ayvaci, A., Angelova, A., Jonschkowski, R.: Smurf: Self- teaching multi-frame unsupervised raft with full-image warping. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 3887–3896 (June 2021)

    2021

  43. [51]

    In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition

    Sui, X., Li, S., Geng, X., Wu, Y., Xu, X., Liu, Y., Goh, R., Zhu, H.: Craft: Cross-attentional flow transformer for robust optical flow. In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition. pp. 17602– 17611 (2022)

    2022

  44. [52]

    In: 2010 IEEE computer society conference on computer vision and pattern recognition

    Sun, D., Roth, S., Black, M.J.: Secrets of optical flow estimation and their princi- ples. In: 2010 IEEE computer society conference on computer vision and pattern recognition. pp. 2432–2439. IEEE (2010)

    2010

  45. [53]

    In: 2012 IEEE Conference on Computer Vision and Pattern Recognition

    Sun, D., Sudderth, E.B., Black, M.J.: Layered segmentation and optical flow es- timation over time. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition. pp. 1768–1775. IEEE (2012)

    2012

  46. [54]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Sun, D., Yang, X., Liu, M.Y., Kautz, J.: Pwc-net: Cnns for optical flow using pyramid, warping, and cost volume. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 8934–8943 (2018)

    2018

  47. [55]

    Advances in Neural Information Processing Systems35, 11313– 11326 (2022)

    Sun, S., Chen, Y., Zhu, Y., Guo, G., Li, G.: Skflow: Learning optical flow with super kernels. Advances in Neural Information Processing Systems35, 11313– 11326 (2022)

    2022

  48. [56]

    Advances in Neural Information Processing Sys- tems31(2018)

    Sundaramoorthi, G., Yezzi, A.: Variational pdes for acceleration on manifolds and application to diffeomorphisms. Advances in Neural Information Processing Sys- tems31(2018)

    2018

  49. [57]

    In: European conference on computer vision

    Teed, Z., Deng, J.: Raft: Recurrent all-pairs field transforms for optical flow. In: European conference on computer vision. pp. 402–419. Springer (2020)

    2020

  50. [58]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Wang, Q., Chang, Y.Y., Cai, R., Li, Z., Hariharan, B., Holynski, A., Snavely, N.: Tracking everything everywhere all at once. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 19795–19806 (2023)

    2023

  51. [59]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Wang, X., Jabri, A., Efros, A.A.: Learning correspondence from the cycle- consistency of time. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 2566–2576 (2019)

    2019

  52. [60]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Wang, Y., Yang, Y., Yang, Z., Zhao, L., Wang, P., Xu, W.: Occlusion aware un- supervised learning of optical flow. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 4884–4893 (2018)

    2018

  53. [61]

    In: European Conference on Computer Vision

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

    2024

  54. [62]

    In: European Conference on Computer Vision

    Wen, H., Liang, E., Deng, J.: Layeredflow: A real-world benchmark for non- lambertian multi-layer optical flow. In: European Conference on Computer Vision. pp. 477–495. Springer (2024)

    2024

  55. [63]

    IEEE Robotics and Automation Letters5(2), 1899–1906 (2020)

    Wong, A., Fei, X., Tsuei, S., Soatto, S.: Unsupervised depth completion from visual inertial odometry. IEEE Robotics and Automation Letters5(2), 1899–1906 (2020)

    1906

  56. [64]

    In: Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition

    Wong, A., Soatto, S.: Bilateral cyclic constraint and adaptive regularization for unsupervised monocular depth prediction. In: Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition. pp. 5644–5653 (2019)

    2019

  57. [65]

    In: Proceedings of the IEEE/CVF International Conference on Com- puter Vision

    Wong, A., Soatto, S.: Unsupervised depth completion with calibrated backprojec- tion layers. In: Proceedings of the IEEE/CVF International Conference on Com- puter Vision. pp. 12747–12756 (2021) 20 Y. Xiao et al

    2021

  58. [66]

    In: European Conference on Computer Vision

    Wu, Y., Liu, T.Y., Park, H., Soatto, S., Lao, D., Wong, A.: Augundo: Scaling up augmentations for monocular depth completion and estimation. In: European Conference on Computer Vision. pp. 274–293. Springer (2024)

    2024

  59. [67]

    IEEE Transactions on Pattern Analysis and Machine Intelligence34(9), 1744–1757 (2011)

    Xu, L., Jia, J., Matsushita, Y.: Motion detail preserving optical flow estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence34(9), 1744–1757 (2011)

    2011

  60. [68]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Yang, Y., Lu, Z., Sundaramoorthi, G.: Coarse-to-fine region selection and match- ing. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 5051–5059 (2015)

    2015

  61. [69]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Yang, Y., Soatto, S.: S2f: Slow-to-fast interpolator flow. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 2087–2096 (2017)

  62. [70]

    In: Proceedings of the European Conference on Computer Vision (ECCV)

    Yang, Y., Soatto, S.: Conditional prior networks for optical flow. In: Proceedings of the European Conference on Computer Vision (ECCV). pp. 271–287 (2018)

    2018

  63. [71]

    IEEE transactions on pattern analysis and machine intelligence37(5), 1053–1066 (2014)

    Yang, Y., Sundaramoorthi, G.: Shape tracking with occlusions via coarse-to-fine region-based sobolev descent. IEEE transactions on pattern analysis and machine intelligence37(5), 1053–1066 (2014)

    2014

  64. [72]

    In: Proceedings of the IEEE International Con- ference on Computer Vision

    Yang, Y., Sundaramoorthi, G., Soatto, S.: Self-occlusions and disocclusions in causal video object segmentation. In: Proceedings of the IEEE International Con- ference on Computer Vision. pp. 4408–4416 (2015)

    2015

  65. [73]

    In: European conference on computer vision

    Yu, J.J., Harley, A.W., Derpanis, K.G.: Back to basics: Unsupervised learning of optical flow via brightness constancy and motion smoothness. In: European conference on computer vision. pp. 3–10. Springer (2016)

    2016

  66. [74]

    ArXivabs/2203.05053(2022),https://api

    Yuan, S., Sun, X., Kim, H., Yu, S., Tomasi, C.: Optical flow training under limited label budget via active learning. ArXivabs/2203.05053(2022),https://api. semanticscholar.org/CorpusID:247362834

    2022 arXiv

  67. [75]

    In: International Conference on Medical Image Computing and Computer-Assisted Intervention

    Zhang, X., Pak, D.H., Ahn, S.S., Li, X., You, C., Staib, L.H., Sinusas, A.J., Wong, A., Duncan, J.S.: Heteroscedastic uncertainty estimation framework for unsuper- vised registration. In: International Conference on Medical Image Computing and Computer-Assisted Intervention. p...

    2024

  68. [76]

    In: European Conference on Computer Vision

    Zhang, X., Stendahl, J.C., Staib, L.H., Sinusas, A.J., Wong, A., Duncan, J.S.: Adaptive correspondence scoring for unsupervised medical image registration. In: European Conference on Computer Vision. pp. 76–92. Springer (2024)

    2024

  69. [77]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Zhao, S., Zhao, L., Zhang, Z., Zhou, E., Metaxas, D.: Global matching with over- lapping attention for optical flow estimation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 17592–17601 (2022)

    2022

  70. [78]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Zhou, T., Jae Lee, Y., Yu, S.X., Efros, A.A.: Flowweb: Joint image set alignment by weaving consistent, pixel-wise correspondences. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 1191–1200 (2015)

    2015

  71. [79]

    In: Proceedings of the European conference on computer vision (ECCV)

    Zou, Y., Luo, Z., Huang, J.B.: Df-net: Unsupervised joint learning of depth and flow using cross-task consistency. In: Proceedings of the European conference on computer vision (ECCV). pp. 36–53 (2018)

    2018

Pith tools

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