pith. sign in

arxiv: 2606.24457 · v1 · pith:TEUB57ZYnew · submitted 2026-06-23 · 💻 cs.CV

Lite Any Stereo V2: Faster and Stronger Efficient Zero-Shot Stereo Matching

Junpeng Jing , Ronglai Zuo , Zhelun Shen , Shangchen Zhou , Rolandos Alexandros Potamias , Stefanos Zafeiriou , Krystian Mikolajczyk , Jiankang Deng This is my paper

Pith reviewed 2026-06-26 00:23 UTC · model grok-4.3

classification 💻 cs.CV
keywords stereo matchingzero-shot generalizationefficient modelscost aggregationknowledge distillationsynthetic-to-real transfercomputer vision
0
0 comments X

The pith

Lite Any Stereo V2 demonstrates that efficient stereo models can exceed the zero-shot accuracy of larger iterative methods while running substantially faster.

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

The paper introduces LAS2, a family of ultra-fast stereo matching models built to deliver strong zero-shot performance without heavy computation or foundation-model priors. It achieves this through a 2D-only cost aggregation architecture tuned for real-world latency and a three-stage training process that moves from synthetic supervision to self-distillation to real-world knowledge distillation. The training includes pseudo-label filtering and error-clamping to make the synthetic-to-real shift more reliable. Experiments show LAS2-H outperforming the iterative Fast-FoundationStereo on overall zero-shot benchmarks while running 1.8 times faster on H200 hardware and 2.7 times faster on Orin hardware.

Core claim

LAS2 revisits efficient stereo design with a 2D-only cost aggregation framework optimized for actual inference latency and applies a three-stage training strategy that combines synthetic supervision, self-distillation, and real-world knowledge distillation. Pseudo-label filtering and an error-clamping operation are added to improve the reliability of the real-world pseudo supervision. This produces a model family whose feed-forward and iterative variants reach state-of-the-art accuracy among efficient stereo methods while maintaining significantly lower latency, with the largest variant delivering stronger overall zero-shot performance than the iterative Fast-FoundationStereo at 1.8x and 2.7

What carries the argument

The 2D-only cost aggregation framework paired with a three-stage training strategy of synthetic supervision, self-distillation, and real-world knowledge distillation that uses pseudo-label filtering and error-clamping.

If this is right

  • Efficient stereo models become viable for deployment on resource-constrained platforms without the accuracy penalty previously assumed.
  • Zero-shot generalization in stereo matching can be improved through careful training progression rather than model scale alone.
  • Trade-offs between speed and accuracy can be managed within a single model family for different hardware budgets.
  • Real-world pseudo supervision can be made more stable by adding filtering and clamping steps during distillation.

Where Pith is reading between the lines

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

  • Similar staged training with filtering may improve zero-shot transfer in related dense prediction tasks such as optical flow.
  • Lower reliance on large foundation models could reduce overall training and inference energy costs for stereo applications.
  • The latency-focused architecture choices may inform efficient designs for other real-time vision systems on edge hardware.

Load-bearing premise

The three-stage training strategy with pseudo-label filtering and error-clamping produces reliable synthetic-to-real transfer without hidden biases or performance drops on unseen real scenes.

What would settle it

Run LAS2-H and Fast-FoundationStereo on a new, previously unseen real-world stereo dataset and measure whether LAS2-H still shows higher overall zero-shot accuracy.

Figures

Figures reproduced from arXiv: 2606.24457 by Jiankang Deng, Junpeng Jing, Krystian Mikolajczyk, Rolandos Alexandros Potamias, Ronglai Zuo, Shangchen Zhou, Stefanos Zafeiriou, Zhelun Shen.

Figure 1
Figure 1. Figure 1: Zero-shot performance on four real-world bench [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Zero-shot prediction on in-the-wild stereo images. We compare disparity maps and reconstructed raw metric point clouds [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Overview of the proposed Lite Any Stereo V2 (LAS2). [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Overview of the proposed three-stage training strategy. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Visualization of validity cues for pseudo-label filtering. The left-right consistency mask, edge mask, and sky segmentation [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Effects of the proposed three-stage training strategy. [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Qualitative comparison with feed-forward stereo methods on in-the-wild stereo images. All methods use the same [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Qualitative comparison with iterative stereo methods on in-the-wild images. The examples include challenging indoor [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Visualization of failure cases. Both the existing method [PITH_FULL_IMAGE:figures/full_fig_p013_9.png] view at source ↗
read the original abstract

Recent advances in stereo matching have achieved remarkable accuracy, but often rely on large models, heavy computation, or additional foundation-model priors, making them difficult to deploy on resource-constrained platforms. In contrast, efficient stereo models offer faster inference but are commonly considered less capable of strong zero-shot generalization. In this paper, we challenge this assumption by introducing Lite Any Stereo V2 (LAS2), an ultra-fast model series designed for efficient zero-shot stereo matching. LAS2 is developed from both architecture and training perspectives. Architecturally, we revisit efficient stereo design under practical deployment settings and propose a 2D-only cost aggregation framework, optimized for real inference latency rather than theoretical MACs alone. For training, we develop a three-stage strategy that combines synthetic supervision, self-distillation, and real-world knowledge distillation. To improve the reliability of real-world pseudo supervision, we further introduce pseudo-label filtering and an error-clamping operation, enabling smoother synthetic-to-real transfer. We instantiate LAS2 as a family of models, including feed-forward variants for different efficiency budgets and an iterative variant for higher accuracy. Extensive experiments show that LAS2 achieves state-of-the-art accuracy among efficient stereo methods while maintaining significantly lower latency. Specifically, LAS2-H achieves stronger overall zero-shot performance than the iterative method Fast-FoundationStereo, with 1.8x and 2.7x faster inference on H200 and Orin, respectively. The project page, demos, and code are available at https://tomtomtommi.github.io/LiteAnyStereoV2/.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 2 minor

Summary. The manuscript introduces Lite Any Stereo V2 (LAS2), a family of efficient models for zero-shot stereo matching. Architecturally, it proposes a 2D-only cost aggregation framework optimized for real inference latency. Training uses a three-stage strategy of synthetic supervision, self-distillation, and real-world knowledge distillation, augmented by pseudo-label filtering and error-clamping to improve synthetic-to-real transfer. Experiments claim LAS2 achieves SOTA accuracy among efficient stereo methods, with the LAS2-H variant outperforming the iterative Fast-FoundationStereo in overall zero-shot performance while delivering 1.8x faster inference on H200 and 2.7x on Orin hardware.

Significance. If the empirical claims hold, the work would be significant for demonstrating that lightweight stereo models can achieve strong zero-shot generalization through targeted training rather than scale or foundation-model priors, with direct relevance to deployment on resource-constrained platforms. The explicit description of the three-stage pipeline, hardware-specific latency results, and code release are positive factors supporting reproducibility and verification.

minor comments (2)
  1. [Abstract] Abstract: the claim of 'stronger overall zero-shot performance' than Fast-FoundationStereo would benefit from a brief parenthetical listing of the primary zero-shot benchmarks (e.g., KITTI, Middlebury) on which the comparison is made.
  2. The manuscript would be strengthened by an explicit statement of the number of runs or error bars for the reported latency and accuracy numbers, even if only in a footnote or appendix.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the constructive review and the recommendation for minor revision. No major comments were raised in the report.

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The paper's derivation consists of an architectural proposal (2D-only cost aggregation optimized for real latency) and an empirical training pipeline (three-stage synthetic supervision + self-distillation + real-world KD with filtering and clamping). All load-bearing claims are measured outcomes on standard zero-shot benchmarks and hardware-specific runtimes, not quantities that reduce to the inputs by definition or by self-citation. No equations or steps equate a claimed prediction to a fitted parameter, and external verification via code release and public datasets keeps the argument self-contained.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard deep-learning assumptions about synthetic data transfer and the effectiveness of distillation; no new mathematical axioms or invented physical entities are introduced.

free parameters (2)
  • model architecture hyperparameters
    Neural network design choices such as layer counts and channel widths are fitted during training on synthetic and real data.
  • pseudo-label filtering thresholds
    Thresholds used to filter real-world pseudo labels are chosen to improve transfer.
axioms (2)
  • domain assumption Synthetic stereo data provides useful initial supervision for real-world generalization
    The first training stage relies on this standard assumption in stereo matching literature.
  • domain assumption Self-distillation and knowledge distillation improve zero-shot performance when combined with filtering
    The three-stage strategy depends on this empirical premise.

pith-pipeline@v0.9.1-grok · 5847 in / 1353 out tokens · 32527 ms · 2026-06-26T00:23:02.172938+00:00 · methodology

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

84 extracted references · 3 linked inside Pith

  1. [1]

    Cooperative computation of stereo disparity,

    D. Marr and T. Poggio, “Cooperative computation of stereo disparity,” inNeurocomputing: foundations of research, 1988, pp. 259–267

    1988

  2. [2]

    Continuous 3d label stereo matching using local expansion moves,

    T. Taniai, Y . Matsushita, Y . Sato, and T. Naemura, “Continuous 3d label stereo matching using local expansion moves,”IEEE TPAMI, vol. 40, no. 11, pp. 2725–2739, 2017

    2017

  3. [3]

    Lite any stereo: Efficient zero-shot stereo matching,

    J. Jing, W. Luo, Y . Mao, and K. Mikolajczyk, “Lite any stereo: Efficient zero-shot stereo matching,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2026, pp. 21 725–21 735

    2026

  4. [4]

    Fast-FoundationStereo: Real-time zero-shot stereo matching,

    B. Wen, S. Dewan, and S. Birchfield, “Fast-FoundationStereo: Real-time zero-shot stereo matching,”CVPR, 2026

    2026

  5. [5]

    Raft-stereo: Multilevel recurrent field transforms for stereo matching,

    L. Lipson, Z. Teed, and J. Deng, “Raft-stereo: Multilevel recurrent field transforms for stereo matching,”arXiv preprint arXiv:2109.07547, 2021

    arXiv 2021

  6. [6]

    Practical stereo matching via cascaded recurrent network with adaptive correlation,

    J. Li, P. Wang, P. Xiong, T. Cai, Z. Yan, L. Yang, J. Liu, H. Fan, and S. Liu, “Practical stereo matching via cascaded recurrent network with adaptive correlation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 16 263–16 272

    2022

  7. [7]

    Accurate and efficient stereo matching via attention concatenation volume,

    G. Xu, Y . Wang, J. Cheng, J. Tang, and X. Yang, “Accurate and efficient stereo matching via attention concatenation volume,”IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

    2023

  8. [8]

    Selective-stereo: Adaptive frequency information selection for stereo matching,

    X. Wang, G. Xu, H. Jia, and X. Yang, “Selective-stereo: Adaptive frequency information selection for stereo matching,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 19 701–19 710

    2024

  9. [9]

    A taxonomy and evaluation of dense two-frame stereo correspondence algorithms,

    D. Scharstein and R. Szeliski, “A taxonomy and evaluation of dense two-frame stereo correspondence algorithms,”IJCV, vol. 47, no. 1, pp. 7–42, 2002

    2002

  10. [10]

    A multi-view stereo benchmark with high- resolution images and multi-camera videos,

    T. Schops, J. L. Schonberger, S. Galliani, T. Sattler, K. Schindler, M. Pollefeys, and A. Geiger, “A multi-view stereo benchmark with high- resolution images and multi-camera videos,” inCVPR, 2017, pp. 3260– 3269

    2017

  11. [11]

    Are we ready for autonomous driving? the kitti vision benchmark suite,

    A. Geiger, P. Lenz, and R. Urtasun, “Are we ready for autonomous driving? the kitti vision benchmark suite,” inCVPR, 2012, pp. 3354– 3361

    2012

  12. [12]

    Object scene flow for autonomous vehicles,

    M. Menze and A. Geiger, “Object scene flow for autonomous vehicles,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2015

    2015

  13. [13]

    Depth anything: Unleashing the power of large-scale unlabeled data,

    L. Yang, B. Kang, Z. Huang, X. Xu, J. Feng, and H. Zhao, “Depth anything: Unleashing the power of large-scale unlabeled data,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 10 371–10 381

    2024

  14. [14]

    Depth anything v2,

    L. Yang, B. Kang, Z. Huang, Z. Zhao, X. Xu, J. Feng, and H. Zhao, “Depth anything v2,”arXiv preprint arXiv:2406.09414, 2024

    Pith/arXiv arXiv 2024

  15. [15]

    Foundationstereo: Zero-shot stereo matching,

    B. Wen, M. Trepte, J. Aribido, J. Kautz, O. Gallo, and S. Birchfield, “Foundationstereo: Zero-shot stereo matching,” 2025. [Online]. Available: https://arxiv.org/abs/2501.09898

    arXiv 2025

  16. [17]

    Defom-stereo: Depth foundation model based stereo matching,

    H. Jiang, Z. Lou, L. Ding, R. Xu, M. Tan, W. Jiang, and R. Huang, “Defom-stereo: Depth foundation model based stereo matching,” 2025. [Online]. Available: https://arxiv.org/abs/2501.09466

    arXiv 2025

  17. [18]

    Mobilestereonet: Towards lightweight deep networks for stereo matching,

    F. Shamsafar, S. Woerz, R. Rahim, and A. Zell, “Mobilestereonet: Towards lightweight deep networks for stereo matching,” inProceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2022, pp. 2417–2426

    2022

  18. [19]

    Lightstereo: Channel boost is all your need for efficient 2d cost aggregation,

    X. Guo, C. Zhang, Y . Zhang, W. Zheng, D. Nie, M. Poggi, and L. Chen, “Lightstereo: Channel boost is all your need for efficient 2d cost aggregation,” 2024. [Online]. Available: https://arxiv.org/abs/2406.19833

    arXiv 2024

  19. [20]

    Banet: Bilateral aggregation network for mobile stereo matching,

    G. Xu, J. Liu, X. Wang, J. Cheng, Y . Deng, J. Zang, Y . Chen, and X. Yang, “Banet: Bilateral aggregation network for mobile stereo matching,”arXiv preprint arXiv:2503.03259, 2025

    arXiv 2025

  20. [21]

    Igev++: Iterative multi-range geometry encoding volumes for stereo matching,

    G. Xu, X. Wang, Z. Zhang, J. Cheng, C. Liao, and X. Yang, “Igev++: Iterative multi-range geometry encoding volumes for stereo matching,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025

    2025

  21. [22]

    Monster++: Unified stereo matching, multi-view stereo, and real-time stereo with monodepth priors,

    J. Cheng, W. Liao, Z. Cai, L. Liu, G. Xu, X. Wang, Y . Wang, Z. Yuan, Y . Deng, J. Zang, Y . Shi, J. Tang, and X. Yang, “Monster++: Unified stereo matching, multi-view stereo, and real-time stereo with monodepth priors,” 2025. [Online]. Available: https://arxiv.org/abs/2501.08643

    arXiv 2025

  22. [23]

    Stereo anything: Unifying stereo matching with large- scale mixed data,

    X. Guo, C. Zhang, Y . Zhang, D. Nie, R. Wang, W. Zheng, M. Poggi, and L. Chen, “Stereo anything: Unifying stereo matching with large- scale mixed data,”arXiv preprint arXiv:2411.14053, 2024

    arXiv 2024

  23. [24]

    A stereo matching algorithm with an adaptive window: Theory and experiment,

    T. Kanade and M. Okutomi, “A stereo matching algorithm with an adaptive window: Theory and experiment,”IEEE transactions on pattern analysis and machine intelligence, vol. 16, no. 9, pp. 920–932, 1994

    1994

  24. [25]

    Performance evaluation of scene registration and stereo matching for artographic feature extrac- tion,

    Y . C. Hsieh, D. M. McKeown, and F. P. Perlant, “Performance evaluation of scene registration and stereo matching for artographic feature extrac- tion,”IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 14, no. 02, pp. 214–238, 1992

    1992

  25. [26]

    Stereo matching with nonlinear diffu- sion,

    D. Scharstein and R. Szeliski, “Stereo matching with nonlinear diffu- sion,”International journal of computer vision, vol. 28, pp. 155–174, 1998

    1998

  26. [27]

    A large dataset to train convolutional networks for disparity, optical flow, and scene flow estimation,

    N. Mayer, E. Ilg, P. Hausser, P. Fischer, D. Cremers, A. Dosovitskiy, and T. Brox, “A large dataset to train convolutional networks for disparity, optical flow, and scene flow estimation,” inCVPR, 2016, pp. 4040–4048. 15

    2016

  27. [28]

    Pyramid stereo matching network,

    J.-R. Chang and Y .-S. Chen, “Pyramid stereo matching network,” in CVPR, 2018, pp. 5410–5418

    2018

  28. [29]

    Group-wise correlation stereo network,

    X. Guo, K. Yang, W. Yang, X. Wang, and H. Li, “Group-wise correlation stereo network,” inCVPR, 2019, pp. 3273–3282

    2019

  29. [30]

    Aanet: Adaptive aggregation network for efficient stereo matching,

    H. Xu and J. Zhang, “Aanet: Adaptive aggregation network for efficient stereo matching,” inCVPR, 2020, pp. 1959–1968

    2020

  30. [31]

    Hitnet: Hierarchical iterative tile refinement network for real-time stereo matching,

    V . Tankovich, C. Hane, Y . Zhang, A. Kowdle, S. Fanello, and S. Bouaziz, “Hitnet: Hierarchical iterative tile refinement network for real-time stereo matching,” inCVPR, 2021, pp. 14 362–14 372

    2021

  31. [32]

    Openstereo: A comprehensive benchmark for stereo matching and strong baseline,

    X. Guo, C. Zhang, J. Lu, Y . Wang, Y . Duan, T. Yang, Z. Zhu, and L. Chen, “Openstereo: A comprehensive benchmark for stereo matching and strong baseline,”arXiv preprint arXiv:2312.00343, 2023

    arXiv 2023

  32. [33]

    Uncertainty guided adaptive warping for robust and efficient stereo matching,

    J. Jing, J. Li, P. Xiong, J. Liu, S. Liu, Y . Guo, X. Deng, M. Xu, L. Jiang, and L. Sigal, “Uncertainty guided adaptive warping for robust and efficient stereo matching,” inProceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), October 2023, pp. 3318–3327

    2023

  33. [34]

    Iterative geometry encoding volume for stereo matching,

    G. Xu, X. Wang, X. Ding, and X. Yang, “Iterative geometry encoding volume for stereo matching,” inProceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, 2023, pp. 21 919– 21 928

    2023

  34. [35]

    Context-enhanced stereo transformer,

    W. Guo, Z. Li, Y . Yang, Z. Wang, R. H. Taylor, M. Unberath, A. Yuille, and Y . Li, “Context-enhanced stereo transformer,” inEuropean Confer- ence on Computer Vision. Springer, 2022, pp. 263–279

    2022

  35. [36]

    Chitransformer: Towards reliable stereo from cues,

    Q. Su and S. Ji, “Chitransformer: Towards reliable stereo from cues,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 1939–1949

    2022

  36. [37]

    Croco v2: Improved cross-view completion pre-training for stereo matching and optical flow,

    P. Weinzaepfel, T. Lucas, V . Leroy, Y . Cabon, V . Arora, R. Br ´egier, G. Csurka, L. Antsfeld, B. Chidlovskii, and J. Revaud, “Croco v2: Improved cross-view completion pre-training for stereo matching and optical flow,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 17 969–17 980

    2023

  37. [38]

    Unifying flow, stereo and depth estimation,

    H. Xu, J. Zhang, J. Cai, H. Rezatofighi, F. Yu, D. Tao, and A. Geiger, “Unifying flow, stereo and depth estimation,”IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023

    2023

  38. [39]

    Dynamicstereo: Consistent dynamic depth from stereo videos,

    N. Karaev, I. Rocco, B. Graham, N. Neverova, A. Vedaldi, and C. Rup- precht, “Dynamicstereo: Consistent dynamic depth from stereo videos,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 13 229–13 239

    2023

  39. [40]

    Match-stereo-videos: Bidirec- tional alignment for consistent dynamic stereo matching,

    J. Jing, Y . Mao, and K. Mikolajczyk, “Match-stereo-videos: Bidirec- tional alignment for consistent dynamic stereo matching,” inEuropean Conference on Computer Vision. Springer, 2024, pp. 415–432

    2024

  40. [41]

    Match stereo videos via bidirectional alignment,

    J. Jing, Y . Mao, A. Qiu, and K. Mikolajczyk, “Match stereo videos via bidirectional alignment,”IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 1–16, 2026

    2026

  41. [42]

    Stereo any video: Tem- porally consistent stereo matching,

    J. Jing, W. Luo, Y . Mao, and K. Mikolajczyk, “Stereo any video: Tem- porally consistent stereo matching,” inProceedings of the IEEE/CVF International Conference on Computer Vision, 2025, pp. 20 836–20 846

    2025

  42. [43]

    Domain- invariant stereo matching networks,

    F. Zhang, X. Qi, R. Yang, V . Prisacariu, B. Wah, and P. Torr, “Domain- invariant stereo matching networks,” inEuropean Conference on Com- puter Vision. Springer, 2020, pp. 420–439

    2020

  43. [44]

    Itsa: An information-theoretic approach to automatic shortcut avoidance and domain generalization in stereo matching networks,

    W. Chuah, R. Tennakoon, R. Hoseinnezhad, A. Bab-Hadiashar, and D. Suter, “Itsa: An information-theoretic approach to automatic shortcut avoidance and domain generalization in stereo matching networks,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 13 022–13 032

    2022

  44. [45]

    Revisiting domain generalized stereo matching networks from a feature consistency perspective,

    J. Zhang, X. Wang, X. Bai, C. Wang, L. Huang, Y . Chen, L. Gu, J. Zhou, T. Harada, and E. R. Hancock, “Revisiting domain generalized stereo matching networks from a feature consistency perspective,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 13 001–13 011

    2022

  45. [46]

    Domain generalized stereo matching via hierarchical visual transformation,

    T. Chang, X. Yang, T. Zhang, and M. Wang, “Domain generalized stereo matching via hierarchical visual transformation,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 9559–9568

    2023

  46. [47]

    Masked representation learning for domain generalized stereo matching,

    Z. Rao, B. Xiong, M. He, Y . Dai, R. He, Z. Shen, and X. Li, “Masked representation learning for domain generalized stereo matching,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023, pp. 5435–5444

    2023

  47. [48]

    Cfnet: Cascade and fused cost volume for robust stereo matching,

    Z. Shen, Y . Dai, and Z. Rao, “Cfnet: Cascade and fused cost volume for robust stereo matching,” inCVPR, 2021, pp. 13 906–13 915

    2021

  48. [49]

    Stereo anywhere: Robust zero-shot deep stereo matching even where either stereo or mono fail,

    L. Bartolomei, F. Tosi, M. Poggi, and S. Mattoccia, “Stereo anywhere: Robust zero-shot deep stereo matching even where either stereo or mono fail,”arXiv preprint arXiv:2412.04472, 2024

    arXiv 2024

  49. [50]

    Learning representa- tions from foundation models for domain generalized stereo matching,

    Y . Zhang, L. Wang, K. Li, Y . Wang, and Y . Guo, “Learning representa- tions from foundation models for domain generalized stereo matching,” inEuropean Conference on Computer Vision. Springer, 2024, pp. 146– 162

    2024

  50. [51]

    All-in-one: Transferring vision foundation models into stereo matching,

    J. Zhou, H. Zhang, J. Yuan, P. Ye, T. Chen, H. Jiang, M. Chen, and Y . Zhang, “All-in-one: Transferring vision foundation models into stereo matching,”arXiv preprint arXiv:2412.09912, 2024

    arXiv 2024

  51. [52]

    Stereonet: Guided hierarchical refinement for real-time edge-aware depth prediction,

    S. Khamis, S. Fanello, C. Rhemann, A. Kowdle, J. Valentin, and S. Izadi, “Stereonet: Guided hierarchical refinement for real-time edge-aware depth prediction,” inECCV, 2018, pp. 573–590

    2018

  52. [53]

    Deeppruner: Learning efficient stereo matching via differentiable patchmatch,

    S. Duggal, S. Wang, W.-C. Ma, R. Hu, and R. Urtasun, “Deeppruner: Learning efficient stereo matching via differentiable patchmatch,” in Proceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 4384–4393

    2019

  53. [54]

    FADNet: A fast and accurate network for disparity estimation,

    Q. Wang, S. Shi, S. Zheng, K. Zhao, and X. Chu, “FADNet: A fast and accurate network for disparity estimation,” in2020 IEEE International Conference on Robotics and Automation (ICRA 2020), 2020, pp. 101– 107

    2020

  54. [55]

    Correlate-and-excite: Real-time stereo matching via guided cost volume excitation,

    A. Bangunharcana, J. W. Cho, S. Lee, I. S. Kweon, K.-S. Kim, and S. Kim, “Correlate-and-excite: Real-time stereo matching via guided cost volume excitation,” in2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2021, pp. 3542–3548

    2021

  55. [56]

    Bilateral grid learning for stereo matching networks,

    B. Xu, Y . Xu, X. Yang, W. Jia, and Y . Guo, “Bilateral grid learning for stereo matching networks,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 12 497–12 506

    2021

  56. [57]

    Attention concatenation volume for accurate and efficient stereo matching,

    G. Xu, J. Cheng, P. Guo, and X. Yang, “Attention concatenation volume for accurate and efficient stereo matching,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 12 981–12 990

    2022

  57. [58]

    Mobilenetv2: Inverted residuals and linear bottlenecks,

    M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, “Mobilenetv2: Inverted residuals and linear bottlenecks,” inProceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 4510–4520

    2018

  58. [59]

    Run, don’t walk: chasing higher flops for faster neural networks,

    J. Chen, S.-h. Kao, H. He, W. Zhuo, S. Wen, C.-H. Lee, and S.-H. G. Chan, “Run, don’t walk: chasing higher flops for faster neural networks,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023, pp. 12 021–12 031

    2023

  59. [60]

    Flickr1024: A large-scale dataset for stereo image super-resolution,

    Y . Wang, L. Wang, J. Yang, W. An, and Y . Guo, “Flickr1024: A large-scale dataset for stereo image super-resolution,” inInternational Conference on Computer Vision Workshops, Oct 2019, pp. 3852–3857

    2019

  60. [61]

    Instereo2k: a large real dataset for stereo matching in indoor scenes,

    W. Bao, W. Wang, Y . Xu, Y . Guo, S. Hong, and X. Zhang, “Instereo2k: a large real dataset for stereo matching in indoor scenes,”Science China Information Sciences, vol. 63, no. 11, pp. 1–11, 2020

    2020

  61. [62]

    Holopix50k: A large-scale in-the-wild stereo image dataset,

    Y . Hua, P. Kohli, P. Uplavikar, A. Ravi, S. Gunaseelan, J. Orozco, and E. Li, “Holopix50k: A large-scale in-the-wild stereo image dataset,” in CVPR Workshop on Computer Vision for Augmented and Virtual Reality, Seattle, WA, 2020., June 2020

    2020

  62. [63]

    Driving- stereo: A large-scale dataset for stereo matching in autonomous driving scenarios,

    G. Yang, X. Song, C. Huang, Z. Deng, J. Shi, and B. Zhou, “Driving- stereo: A large-scale dataset for stereo matching in autonomous driving scenarios,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 899–908

    2019

  63. [64]

    Uasol, a large-scale high-resolution outdoor stereo dataset,

    Z. Bauer, F. Gomez-Donoso, E. Cruz, S. Orts-Escolano, and M. Cazorla, “Uasol, a large-scale high-resolution outdoor stereo dataset,”Scientific data, vol. 6, no. 1, p. 162, 2019

    2019

  64. [65]

    Falling things: A synthetic dataset for 3d object detection and pose estimation,

    J. Tremblay, T. To, and S. Birchfield, “Falling things: A synthetic dataset for 3d object detection and pose estimation,” inCVPRW, 2018, pp. 2038–2041

    2018

  65. [66]

    Virtual kitti 2,

    Y . Cabon, N. Murray, and M. Humenberger, “Virtual kitti 2,” 2020

    2020

  66. [67]

    Tartanair: A dataset to push the limits of visual slam,

    W. Wang, D. Zhu, X. Wang, Y . Hu, Y . Qiu, C. Wang, Y . Hu, A. Kapoor, and S. Scherer, “Tartanair: A dataset to push the limits of visual slam,” 2020

    2020

  67. [68]

    Irs: A large naturalistic indoor robotics stereo dataset to train deep models for dis- parity and surface normal estimation,

    Q. Wang, S. Zheng, Q. Yan, F. Deng, K. Zhao, and X. Chu, “Irs: A large naturalistic indoor robotics stereo dataset to train deep models for dis- parity and surface normal estimation,”arXiv preprint arXiv:1912.09678, 2019

    arXiv 1912

  68. [69]

    A naturalistic open source movie for optical flow evaluation,

    D. J. Butler, J. Wulff, G. B. Stanley, and M. J. Black, “A naturalistic open source movie for optical flow evaluation,” inECCV, 2012, pp. 611–625

    2012

  69. [70]

    Spring: A high-resolution high-detail dataset and benchmark for scene flow, optical flow and stereo,

    L. Mehl, J. Schmalfuss, A. Jahedi, Y . Nalivayko, and A. Bruhn, “Spring: A high-resolution high-detail dataset and benchmark for scene flow, optical flow and stereo,” inProc. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023

    2023

  70. [71]

    Promptstereo: Zero-shot stereo matching via structure and motion prompts,

    X. Wang, H. Yang, H. Wang, J. Cheng, G. Xu, M. Lin, and X. Yang, “Promptstereo: Zero-shot stereo matching via structure and motion prompts,”arXiv preprint arXiv:2603.01650, 2026

    arXiv 2026

  71. [72]

    Acceleration of stochastic approxima- tion by averaging,

    B. T. Polyak and A. B. Juditsky, “Acceleration of stochastic approxima- tion by averaging,”SIAM journal on control and optimization, vol. 30, no. 4, pp. 838–855, 1992

    1992

  72. [73]

    Sam 3: Segment anything with concepts,

    N. Carion, L. Gustafson, Y .-T. Hu, S. Debnath, R. Hu, D. Suris, C. Ryali, K. V . Alwala, H. Khedr, A. Huanget al., “Sam 3: Segment anything with concepts,”arXiv preprint arXiv:2511.16719, 2025. 16

    Pith/arXiv arXiv 2025

  73. [74]

    Stereo4d: Learning how things move in 3d from internet stereo videos,

    L. Jin, R. Tucker, Z. Li, D. Fouhey, N. Snavely, and A. Holynski, “Stereo4d: Learning how things move in 3d from internet stereo videos,” arXiv preprint, 2024

    2024

  74. [75]

    Structure- guided ranking loss for single image depth prediction,

    K. Xian, J. Zhang, O. Wang, L. Mai, Z. Lin, and Z. Cao, “Structure- guided ranking loss for single image depth prediction,” inThe IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2020

    2020

  75. [76]

    Confidence aware stereo matching for realistic cluttered scenario,

    J. Min and Y . Jeon, “Confidence aware stereo matching for realistic cluttered scenario,” in2024 IEEE International Conference on Image Processing (ICIP). IEEE, 2024, pp. 3491–3497

    2024

  76. [77]

    Adam: A method for stochastic optimization,

    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, 2014

    Pith/arXiv arXiv 2014

  77. [78]

    A convnet for the 2020s,

    Z. Liu, H. Mao, C.-Y . Wu, C. Feichtenhofer, T. Darrell, and S. Xie, “A convnet for the 2020s,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 11 976–11 986

    2022

  78. [79]

    Searching for mobilenetv3,

    A. Howard, M. Sandler, G. Chu, L.-C. Chen, B. Chen, M. Tan, W. Wang, Y . Zhu, R. Pang, V . Vasudevanet al., “Searching for mobilenetv3,” inProceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 1314–1324

    2019

  79. [80]

    Efficientnetv2: Smaller models and faster training,

    M. Tan and Q. Le, “Efficientnetv2: Smaller models and faster training,” inInternational conference on machine learning. PMLR, 2021, pp. 10 096–10 106

    2021

  80. [81]

    Ghostnet: More features from cheap operations,

    K. Han, Y . Wang, Q. Tian, J. Guo, C. Xu, and C. Xu, “Ghostnet: More features from cheap operations,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2020, pp. 1580– 1589

    2020

Showing first 80 references.