Pith. sign in

REVIEW 2 major objections 5 minor 1 cited by

FlowMamba: Learning Point Cloud Scene Flow with Global Motion Propagation

T0 review · 2 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read FlowMamba claims that replacing local recurrent updates with a bidirectional state-space model guided by a learned point ordering cuts point-cloud scene flow error by roughly a fifth and reaches sub-centimeter accuracy on standard…

desk verdict FlowMamba delivers strong scene flow gains via bidirectional Mamba, but the FIO ordering's gradient path is unspecified and Table 3 has a clear number bug. read the letter →

arxiv 2412.17366 v1 pith:YIXI5K3P submitted 2024-12-23 cs.CV

Min Lin , Gangwei Xu , Yun Wang , Xianqi Wang , Xin Yang This is my paper · ORCID
classification cs.CV
keywords sceneflowpointcloudstatespacemodelMambaglobalmotionpropagationiterativeupdatelearnedorderingill-posedregions
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proposes FlowMamba, a network that estimates point-cloud scene flow—the 3D motion of each point between two frames—by propagating global motion information through a bidirectional state space model instead of relying only on local neighborhoods. The central claim is that its Iterative SSM-based Update module (ISU) resolves the ambiguity of flat, occluded, and slender regions where local evidence is insufficient, and that its Feature-Induced Ordering (FIO) makes the point sequence spatially coherent so the SSM can actually propagate information. On FlyingThings3D and KITTI the method reports EPE3D of 0.0089 and 0.0062 meters, reductions of 21.9% and 20.5% over DifFlow3D, and the first millimeter-level accuracy on both benchmarks. The authors also claim the ISU is a plug-and-play replacement for GRU update cells in existing iterative scene flow networks, improving their accuracy as well.

What carries the argument

Two components carry the argument. The ISU (Iterative SSM-based Update) takes context, motion, and hidden features, orders the points according to a learned score, applies several bidirectional Mamba blocks to the sorted sequence, and adaptively fuses the output with the previous hidden state via a sigmoid gate—this is what propagates global motion patterns. The FIO (Feature-Induced Ordering) computes Score = tanh(MLP[cfl, mfl,n, hl,n-1]) for each point and sorts all points by that score, aiming to turn the irregular point cloud into a sequence with spatial continuity so that the SSM's causal dependencies are meaningful.

What would settle it

Train FlowMamba with FIO replaced by a fixed, non-learned ordering such as lexicographic sorting by 3D coordinates, or a random permutation, keeping all else fixed; if the EPE3D on FlyingThings3D remains at 0.0089 meters, the learned ordering is not the source of the improvement.

Watch

Extended reading notes

Core claim

The paper's core discovery, on its own terms, is that scene flow accuracy in ill-posed regions is limited by the local receptive field of the recurrent update cell, and that a bidirectional structured state space model with a learned ordering can supply the missing global context. FlowMamba replaces the GRU in a coarse-to-fine MSBRN-style architecture with an ISU that sorts points by a learned score, passes hidden states through bidirectional Mamba blocks, and fuses the globally refined hidden state with the previous iteration's state using a learned gate. This combination achieves state-of-the-art end-point error on both the synthetic FlyingThings3D and real-world KITTI benchmarks without fine-tuning on KITTI, and the ablation table attributes the gain specifically to the ISU and FIO components.

Load-bearing premise

The ordering is claimed to be learned, but the paper does not explain how gradients pass through the discrete sorting step to train the MLP that produces the scores, so the FIO may not actually be learning anything.

Editorial extensions

If this is right

  • If correct, global sequence-based motion propagation is a viable fix for ill-posed scene flow regions, not just local cost volumes.
  • Iterative scene flow methods can drop in the ISU in place of GRU and gain accuracy without retraining from scratch.
  • Millimeter-level accuracy on KITTI without fine-tuning implies synthetic-to-real generalization is strong enough for practical deployment.
  • Mamba's linear complexity makes the global receptive field affordable on 8k-point clouds.
  • Fewer iterations reach baseline accuracy, suggesting faster inference options.

Reading between the lines

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

  • We infer, though the paper does not test it, that the FIO style of learned ordering could transfer to other point-cloud sequence tasks such as object segmentation or completion, where spatial continuity in a scan order may similarly help SSMs.
  • An implication left implicit is that any cheap spatial ordering that already encodes locality might capture part of FIO's gain; ablating against such orderings would isolate what the learned order adds.
  • The claim that ISU propagates global information could be tested directly by probing how much the hidden state of a distant point changes after one ISU pass, a diagnostic the paper does not report.
  • If the ordering is not trainable end-to-end, the observed gains might come from the Bi-Mamba block alone on a fixed order; the paper's own unidirectional-Mamba ablation suggests directionality matters more than the score.
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

2 major / 5 minor

Summary. FlowMamba proposes a supervised point-cloud scene-flow network built on the MSBRN coarse-to-fine backbone, replacing the GRU update with an Iterative SSM-based Update (ISU) module that uses bidirectional Mamba blocks to propagate global motion information, and adding a Feature-Induced Ordering (FIO) strategy that sorts points by a learned score before sequence modeling. On FlyingThings3D and KITTI the method reports state-of-the-art EPE3D values (0.0089 and 0.0062 m, reductions of 21.9% and 20.5% over DifFlow3D), with additional experiments showing that ISU improves existing iterative methods and ablations of the ISU and FIO components.

Significance. If the results are reproducible, the paper makes a solid empirical contribution: it is, to my knowledge, the first application of bidirectional Mamba/SSM sequence modeling to point-cloud scene flow, and the reported gains are large and consistent across both synthetic and real-world benchmarks. The paper includes extensive comparisons (Tables 1 and 2), ablations (Tables 4 to 6), a runtime comparison (Table 7), and a plug-and-play study (Table 3). The arithmetic in the headline EPE3D reductions checks out. The two issues below -- the unspecified gradient path for the learned ordering and the inconsistent Outliers values in Table 3 -- are load-bearing and need to be fixed before the claims can be fully trusted.

major comments (2)
  1. [Section 3, 'Feature-induced Ordering Strategy', Eq. (6)] The FIO strategy computes Score = tanh(MLP([cfl, mfl,n, hl,n-1])) and then sorts all points by this score to define the sequence fed into the bidirectional Mamba blocks. Sorting is a discrete, non-differentiable operation: a naive implementation (e.g., torch.argsort) gives zero gradient to the score MLP. The paper does not describe any differentiable surrogate such as a soft sort, a perturbed optimizer, or a straight-through estimator, nor does it state that gradients are stopped or that the permutation is treated as a fixed preprocessing step. This matters because the FIO ablations in Table 5 are interpreted as evidence that the ordering is learned from the three features; with an untrained score MLP, the differences between ablations would reflect fixed random permutations rather than a learned feature-induced ordering. The authors must specify the exact gradient path used in training, or present evidence (e.g., before/after training analysis of the scores) that the ordering is actually learned.
  2. [Table 3 (Universality of proposed modules)] The reported Outliers value for MSBRN on FlyingThings3D, 0.0243, duplicates the Ours value in Table 1 for the same metric and contradicts Table 1's MSBRN Outliers value of 0.0560. Moreover, MSBRN(w/ISU) shows Outliers 0.0892 on FlyingThings3D, which is worse than both possible baselines (0.0560 or 0.0243), even though the same row reports a 28.6% EPE3D improvement. This inconsistency directly undermines the claim that ISU improves baseline accuracy across all metrics. Please correct the table and re-evaluate the plug-and-play claim for the Outliers metric on FlyingThings3D; if the improvement is not consistent, the claim should be qualified accordingly.
minor comments (5)
  1. [Conclusion] The conclusion states that the proposed method shows 'a strong university'; this should read 'universality'.
  2. [Preliminaries, Eq. (2)] In Eq. (2), the symbol B is used both for the continuous projection matrix and for the discretized matrix; using \bar{B} or another symbol for the discretized version would avoid confusion.
  3. [Abstract and Introduction] There are several grammatical issues, for example 'There has been a significantly increasing focus' and 'the same number of iterations as reporting in (Cheng and Ko 2023)'; these should be fixed in a polish pass.
  4. [Table 5] The FIO ablation differences are small (EPE3D ranging from 0.0096 to 0.0089); reporting error bars or multiple seeds would make the conclusions more robust.
  5. [Implementation Details] Since the FIO gradient path is central to the method, a short pseudocode block or a statement about the differentiable sorting mechanism would improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: FlowMamba's SOTA claims are supervised benchmark evaluations against external baselines; the FIO differentiability concern is an implementation/specification issue, not a circular derivation.

full rationale

FlowMamba's derivation chain is self-contained in the relevant sense: the central claims are empirical state-of-the-art comparisons and ablations on held-out FlyingThings3D and KITTI test sets against external published baselines (DifFlow3D, MSBRN, PV-RAFT, PT-FlowNet, etc.), and the ISU/FIO modules are trained by a supervised L2 loss against ground-truth flow (Eq. 7). No fitted parameter is relabeled as a prediction, and no load-bearing premise is justified solely by a self-citation: references to the authors' earlier papers appear only as related work or component-level citations, while the architectural template (MSBRN) and the SSM machinery (Mamba) are external and reproducible. The concern that the sorting step in Eq. (6) is non-differentiable and may leave the FIO score MLP untrained is a correctness/implementation caveat about whether the claimed causal mechanism is actually learned; it does not reduce any claimed result to its own input, and even if true it would weaken the FIO interpretation without making the benchmark evaluation circular. No specific reduction of the kind required for a circularity finding can be exhibited from the paper's equations or citation chain.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on standard supervised learning assumptions and on the two proposed architectural components, ISU and FIO, which are not physical entities. The main free parameters are training choices and the unspecified architecture dimensions.

free parameters (3)
  • alpha(l) loss weights = [0.16, 0.08, 0.04, 0.02]
    Adopted from prior work (Cheng and Ko 2023; Liu et al. 2023); the total loss in Eq. (7) depends on them.
  • Number of update iterations per layer = 4 during training and evaluation
    Chosen by hand; Table 6 shows it trades off accuracy and runtime.
  • Architecture dimensions = Not reported
    Hidden width C, number of Bi-Mamba blocks, SSM state size, and FIO MLP size are not specified, so the trained network is not fully defined.
assumptions (4)
  • domain assumption FlyingThings3D ground-truth scene flow labels are accurate enough for supervised training.
    The loss in Eq. (7) treats s_f^gt as ground truth; if labels are noisy, reported errors are biased.
  • domain assumption Zero-shot evaluation on KITTI (without fine-tuning) is a valid measure of real-world generalization.
    The paper reports KITTI numbers from a model trained only on synthetic data; this assumes the train/test distribution shift is representative.
  • domain assumption Uniform random sampling of 8192 points preserves the motion field and correspondences.
    Sampling is used for all comparisons; if sampling changes the difficulty, fair comparison could be affected.
  • standard math The zero-order hold discretization of the continuous SSM (Eq. 2) is valid.
    The Mamba backbone relies on this standard discretization; the paper introduces no modification.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FlowMamba: Learning Point Cloud Scene Flow with Global Motion Propagation." pith.science (2026). https://pith.science/paper/YIXI5K3P

@misc{pith2026241217366,
  author       = {Pith},
  title        = {Pith review of: FlowMamba: Learning Point Cloud Scene Flow with Global Motion Propagation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YIXI5K3P}},
  note         = {Machine review of arXiv:2412.17366}
}
read the original abstract

Scene flow methods based on deep learning have achieved impressive performance. However, current top-performing methods still struggle with ill-posed regions, such as extensive flat regions or occlusions, due to insufficient local evidence. In this paper, we propose a novel global-aware scene flow estimation network with global motion propagation, named FlowMamba. The core idea of FlowMamba is a novel Iterative Unit based on the State Space Model (ISU), which first propagates global motion patterns and then adaptively integrates the global motion information with previously hidden states. As the irregular nature of point clouds limits the performance of ISU in global motion propagation, we propose a feature-induced ordering strategy (FIO). The FIO leverages semantic-related and motion-related features to order points into a sequence characterized by spatial continuity. Extensive experiments demonstrate the effectiveness of FlowMamba, with 21.9\% and 20.5\% EPE3D reduction from the best published results on FlyingThings3D and KITTI datasets. Specifically, our FlowMamba is the first method to achieve millimeter-level prediction accuracy in FlyingThings3D and KITTI. Furthermore, the proposed ISU can be seamlessly embedded into existing iterative networks as a plug-and-play module, improving their estimation accuracy significantly.

Figures

Figures reproduced from arXiv: 2412.17366 by the authors.

Figure 1
Figure 1. Left: Comparison with state-of-the-art scene flow methods (Liu et al. 2024a; Cheng and Ko 2023; Wang et al. 2023a; Fu et al. 2023; Cheng and Ko 2022; Wang et al. 2022) on FlyingThings3D and KITTI. Notably, we achieved millimeter-level precision on both datasets for the first time. Right: Comparison with the accuracy of each layer output on FlyingThings3D. Our FlowMamba can achieve superior results from the coarsest … view at source ↗
Figure 2
Figure 2. Overview of our proposed FlowMamba. Feature encoders abstract the point clouds to obtain the multi-scale point [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The architecture of proposed module. Left: Iterative SSM-based Update (ISU) Module. Middle: Feature-Induced [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Qualitative results on the test set of KITTI. It shows that providing global motion propagation improves performance in areas with ambiguous geometric characteristics, such as embankments, roadside grassy areas, and some slender structures (curbs or tracks). Blue, gree…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. A Survey on Mamba Architecture for Vision Applications

    cs.CV 2025-02 conditional novelty 1.0 of 10

    A survey of Mamba-based vision models that summarizes scanning mechanisms, key architectures, and benchmark results, contributing no new experimental findings.

Reference graph

Works this paper leans on

50 extracted references · 28 canonical work pages · cited by 1 Pith paper

  1. [1]

    N.; and Stricker, D

    Battrawy, R.; Schuster, R.; Mahani, M.-A. N.; and Stricker, D. 2022. Rms-flownet: Efficient and robust multi-scale scene flow estimation for large-scale point clouds. In 2022 International Conference on Robotics and Automation (ICRA), 883--889. IEEE

    work page 2022

  2. [2]

    Cheng, W.; and Ko, J. H. 2022. Bi-pointflownet: Bidirectional learning for point cloud based scene flow estimation. In European Conference on Computer Vision, 108--124. Springer

    work page 2022

  3. [3]

    Cheng, W.; and Ko, J. H. 2023. Multi-scale bidirectional recurrent network with hybrid correlation for point cloud based scene flow estimation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 10041--10050

    work page 2023

  4. [4]

    Chollet, F. 2017. Xception: Deep learning with depthwise separable convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition, 1251--1258

    2017

  5. [5]

    Fu, J.; Xiang, Z.; Qiao, C.; and Bai, T. 2023. PT-FlowNet: Scene Flow Estimation on Point Clouds With Point Transformer. IEEE Robotics and Automation Letters, 8(5): 2566--2573

    work page 2023

  6. [6]

    Geiger, A.; Lenz, P.; Stiller, C.; and Urtasun, R. 2013. Vision meets robotics: The kitti dataset. The International Journal of Robotics Research, 32(11): 1231--1237

    2013

  7. [7]

    Gu, A.; and Dao, T. 2023. Mamba: Linear-time sequence modeling with selective state spaces. arXiv preprint arXiv:2312.00752

    arXiv 2023

  8. [8]

    Gu, A.; Goel, K.; and R \'e , C. 2021. Efficiently modeling long sequences with structured state spaces. arXiv preprint arXiv:2111.00396

    arXiv 2021

Show all 50 references
  1. [9]

    Gu, X.; Tang, C.; Yuan, W.; Dai, Z.; Zhu, S.; and Tan, P. 2022. Rcp: Recurrent closest point for point cloud. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8216--8226

    2022

  2. [10]

    J.; and Wang, P

    Gu, X.; Wang, Y.; Wu, C.; Lee, Y. J.; and Wang, P. 2019. Hplflownet: Hierarchical permutohedral lattice flownet for scene flow estimation on large-scale point clouds. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 3254--3263

    2019

  3. [11]

    Hendrycks, D.; and Gimpel, K. 2016. Gaussian error linear units (gelus). arXiv preprint arXiv:1606.08415

    2016 arXiv

  4. [12]

    P.; and Ba, J

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

    2014 arXiv

  5. [13]

    C.; and Raviv, D

    Kittenplon, Y.; Eldar, Y. C.; and Raviv, D. 2021. Flowstep3d: Model unrolling for self-supervised scene flow estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4114--4123

    2021

  6. [14]

    Li, R.; Lin, G.; He, T.; Liu, F.; and Shen, C. 2021. Hcrf-flow: Scene flow from point clouds with continuous high-order crfs and position-aware flow embedding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 364--373

    2021

  7. [15]

    Li, Y.; Yang, W.; and Fei, B. 2024. 3dmambacomplete: Exploring structured state space model for point cloud completion. arXiv preprint arXiv:2404.07106

    2024 arXiv

  8. [16]

    Liang, D.; Zhou, X.; Wang, X.; Zhu, X.; Xu, W.; Zou, Z.; Ye, X.; and Bai, X. 2024. Pointmamba: A simple state space model for point cloud analysis. arXiv preprint arXiv:2402.10739

    2024 arXiv

  9. [17]

    Liu, J.; Wang, G.; Ye, W.; Jiang, C.; Han, J.; Liu, Z.; Zhang, G.; Du, D.; and Wang, H. 2023. DifFlow3D: Toward Robust Uncertainty-Aware Scene Flow Estimation with Diffusion Model. arXiv preprint arXiv:2311.17456

    2023 arXiv

  10. [18]

    Liu, J.; Wang, G.; Ye, W.; Jiang, C.; Han, J.; Liu, Z.; Zhang, G.; Du, D.; and Wang, H. 2024 a . DifFlow3D: Toward Robust Uncertainty-Aware Scene Flow Estimation with Iterative Diffusion-Based Refinement. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern...

    2024

  11. [19]

    Liu, J.; Yu, R.; Wang, Y.; Zheng, Y.; Deng, T.; Ye, W.; and Wang, H. 2024 b . Point mamba: A novel point cloud backbone based on state space model with octree-based ordering strategy. arXiv preprint arXiv:2403.06467

    2024 arXiv

  12. [20]

    R.; and Guibas, L

    Liu, X.; Qi, C. R.; and Guibas, L. J. 2019. Flownet3d: Learning scene flow in 3d point clouds. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 529--537

    2019

  13. [21]

    Loshchilov, I.; and Hutter, F. 2017. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101

    2017 arXiv

  14. [22]

    Lu, Z.; and Cheng, M. 2023. GMA3D: Local-global attention learning to estimate occluded motions of scene flow. In Chinese Conference on Pattern Recognition and Computer Vision (PRCV), 16--27. Springer

    2023

  15. [23]

    Mayer, N.; Ilg, E.; Hausser, P.; Fischer, P.; Cremers, D.; Dosovitskiy, A.; and Brox, T. 2016. 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 recogniti...

    2016

  16. [24]

    Mo, S. 2024. Efficient 3D Shape Generation via Diffusion Mamba with Bidirectional SSMs. arXiv preprint arXiv:2406.05038

    2024 arXiv

  17. [25]

    Paszke, A.; Gross, S.; Massa, F.; Lerer, A.; Bradbury, J.; Chanan, G.; Killeen, T.; Lin, Z.; Gimelshein, N.; Antiga, L.; et al. 2019. Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems, 32

    2019

  18. [26]

    R.; Su, H.; Mo, K.; and Guibas, L

    Qi, C. R.; Su, H.; Mo, K.; and Guibas, L. J. 2017 a . Pointnet: Deep learning on point sets for 3d classification and segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition, 652--660

    2017

  19. [27]

    R.; Yi, L.; Su, H.; and Guibas, L

    Qi, C. R.; Yi, L.; Su, H.; and Guibas, L. J. 2017 b . Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Advances in neural information processing systems, 30

    2017

  20. [28]

    J.; Li, E

    Seita, D.; Wang, Y.; Shetty, S. J.; Li, E. Y.; Erickson, Z.; and Held, D. 2023. Toolflownet: Robotic manipulation with tools via predicting tool flow from point clouds. In Conference on Robot Learning, 1038--1049. PMLR

    2023

  21. [29]

    Teng, S.; Hu, X.; Deng, P.; Li, B.; Li, Y.; Ai, Y.; Yang, D.; Li, L.; Xuanyuan, Z.; Zhu, F.; et al. 2023. Motion planning for autonomous driving: The state of the art and future perspectives. IEEE Transactions on Intelligent Vehicles

    2023

  22. [30]

    Wang, G.; Hu, Y.; Liu, Z.; Zhou, Y.; Tomizuka, M.; Zhan, W.; and Wang, H. 2022. What matters for 3d scene flow network. In European Conference on Computer Vision, 38--55. Springer

    2022

  23. [31]

    Wang, G.; Wu, X.; Liu, Z.; and Wang, H. 2021. Hierarchical attention learning of scene flow in 3d point clouds. IEEE Transactions on Image Processing, 30: 5168--5181

    2021

  24. [32]

    Wang, X.; Xu, G.; Jia, H.; and Yang, X. 2024. Selective-stereo: Adaptive frequency information selection for stereo matching. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 19701--19710

    2024

  25. [33]

    Wang, Y.; Chi, C.; Lin, M.; and Yang, X. 2023 a . Ihnet: Iterative hierarchical network guided by high-resolution estimated information for scene flow estimation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 10073--10082

    2023

  26. [34]

    Wang, Z.; Nguyen, C.; Asente, P.; and Dorsey, J. 2023 b . PointShopAR: Supporting Environmental Design Prototyping Using Point Cloud in Augmented Reality. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, 1--15

    2023

  27. [35]

    Wei, Y.; Wang, Z.; Rao, Y.; Lu, J.; and Zhou, J. 2021. Pv-raft: Point-voxel correlation fields for scene flow estimation of point clouds. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 6954--6963

    2021

  28. [36]

    Wu, W.; Fuxin, L.; and Shan, Q. 2023. Pointconvformer: Revenge of the point-based convolution. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 21802--21813

    2023

  29. [37]

    Wu, W.; Qi, Z.; and Fuxin, L. 2019. Pointconv: Deep convolutional networks on 3d point clouds. In Proceedings of the IEEE/CVF Conference on computer vision and pattern recognition, 9621--9630

    2019

  30. [38]

    Y.; Li, Z.; Liu, W.; and Fuxin, L

    Wu, W.; Wang, Z. Y.; Li, Z.; Liu, W.; and Fuxin, L. 2020. Pointpwc-net: Cost volume on point clouds for (self-) supervised scene flow estimation. In Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part V 16, 88--107. Springer

    2020

  31. [39]

    Xu, G.; Cheng, J.; Guo, P.; and Yang, X. 2022. Attention concatenation volume for accurate and efficient stereo matching. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 12981--12990

    2022

  32. [40]

    Xu, G.; Wang, X.; Ding, X.; and Yang, X. 2023 a . Iterative geometry encoding volume for stereo matching. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 21919--21928

    2023

  33. [41]

    Xu, G.; Wang, X.; Zhang, Z.; Cheng, J.; Liao, C.; and Yang, X. 2024 a . Igev++: Iterative multi-range geometry encoding volumes for stereo matching. arXiv preprint arXiv:2409.00638

    2024 arXiv

  34. [42]

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

    2023

  35. [43]

    Xu, G.; Wang, Y.; Gu, J.; Xue, T.; and Yang, X. 2024 b . HDRFlow: Real-Time HDR Video Reconstruction with Large Motions. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 24851--24860

    2024

  36. [44]

    Zhang, D.; Zhang, H.; Tang, J.; Wang, M.; Hua, X.; and Sun, Q. 2020. Feature pyramid transformer. In Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part XXVIII 16, 323--339. Springer

    2020

  37. [45]

    Zhang, T.; Li, X.; Yuan, H.; Ji, S.; and Yan, S. 2024 a . Point could mamba: Point cloud learning via state space model. arXiv preprint arXiv:2403.00762

    2024 arXiv

  38. [46]

    Zhang, Y.; Edstedt, J.; Wandt, B.; Forss \'e n, P.-E.; Magnusson, M.; and Felsberg, M. 2024 b . Gmsf: Global matching scene flow. Advances in Neural Information Processing Systems, 36

    2024

  39. [47]

    H.; and Koltun, V

    Zhao, H.; Jiang, L.; Jia, J.; Torr, P. H.; and Koltun, V. 2021. Point transformer. In Proceedings of the IEEE/CVF international conference on computer vision, 16259--16268

    2021

  40. [48]

    Zhu, L.; Liao, B.; Zhang, Q.; Wang, X.; Liu, W.; and Wang, X. 2024. Vision mamba: Efficient visual representation learning with bidirectional state space model. arXiv preprint arXiv:2401.09417

    2024 arXiv

  41. [49]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  42. [50]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

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