Pith. sign in

REVIEW 4 major objections 4 minor 1 cited by

Robust Monocular Visual Odometry using Curriculum Learning

T0 review · 4 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A training-only curriculum over DPVO's losses beats prior monocular visual odometry, cutting average trajectory error from 0.21 m to 0.14 m.

desk verdict Useful empirical study of curriculum learning for DPVO, but the headline self-paced variant has an equations-level bug that undercuts the claimed mechanism; the trajectory and DDPG variants are more credible. read the letter →

arxiv 2411.13438 v2 pith:Y5YZNNHM submitted 2024-11-20 cs.RO cs.AI

Assaf Lahiany , Oren Gal This is my paper · ORCID
classification cs.ROcs.AI
keywords monocularvisualodometrycurriculumlearningself-paceddeeppatchTartanAirreinforcementabsolutetrajectoryerrorsim-to-realtransfer
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

The paper sets out to show that the accuracy and robustness of monocular visual odometry can be improved by changing how the DPVO network is trained, not by redesigning its architecture or adding sensors. It wraps the DPVO loss in a hierarchical curriculum that progressively weights flow, translation, and rotation terms, and compares three schedulers: trajectory-based difficulty phases, self-paced exponential weighting, and DDPG agents. In the authors' evaluation the self-paced variant reaches an average ATE of 0.14 m on the TartanAir test split, a 33% relative improvement over DPVO's 0.21 m and an 18% improvement over the image-event method RAMP-VO. If this holds, a training-only modification can lift a leading VO system past the state of the art and make it more consistent on hard motion.

What carries the argument

The load-bearing object is a hierarchical curriculum loss $L_{\text{total}} = w_f s_f L_{\text{flow}} + w_p s_p (L_{\text{translation}} + w_r L_{\text{rotation}})$ with fixed base scales $s_f=0.1$ and $s_p=10$, plus three dynamic weights $w_f, w_p, w_r$. Three schedulers drive those weights: trajectory-based phases built from normalized six-degree-of-freedom motion scores; a self-paced factor $\varphi(L_i)=e^{-\lambda L_i}$; and DDPG agents that output continuous actions per component with negative component loss as reward. The self-paced factor is the mechanism behind the main SOTA claim, since it smoothly ramps all curriculum weights as the total loss falls.

What would settle it

A controlled experiment that trains DPVO with the same self-paced weight values but on a fixed step-count schedule, removing the dependence of $\varphi(L_i)$ on the current loss, would settle the mechanism; if average ATE on TartanAir stays at 0.14 m, loss-driven curriculum ordering is not what produces the gain.

Watch

Extended reading notes

Core claim

The paper's central claim is that curriculum learning over the existing DPVO training objective yields state-of-the-art monocular visual odometry without architectural changes. The self-paced variant defines a progress factor $\varphi(L_i)=e^{-\lambda L_i}$ from the current total loss and uses it to interpolate the flow, pose, and rotation weights from $w_0=0.1$ to $w_F=1$, so early high losses suppress hard examples and lower losses let the weights grow. In the paper's experiments this reaches AUC 0.87 and median ATE 0.14 m on TartanAir, cutting average ATE by 33% relative to DPVO and by 18% relative to RAMP-VO, with the lowest global ATE standard deviation. The same model reduces average ATE on EuRoC by 13% and on TUM-RGBD by 9%, and it reaches DPVO-level validation performance with about 47% fewer training steps.

Load-bearing premise

The load-bearing premise is that the self-paced factor defined in Section III.D.2, computed from the total loss and applied identically to every loss weight, actually creates a curriculum over sample difficulty; if it only scales the whole loss up and down over time, the reported gains could come from an adaptive loss-scale effect rather than an easy-to-hard ordering.

Editorial extensions

If this is right

  • A training-only change can push a strong patch-based VO system past image-event baselines, so further VO accuracy gains may not require new sensors or heavier inference.
  • The self-paced schedule reaches DPVO-level validation performance in about 47% fewer training steps, reducing the compute cost of training on large synthetic datasets.
  • Synthetic-to-real transfer improves on EuRoC and TUM-RGBD, meaning the curriculum helps the model generalize beyond the motion and appearance distribution it saw in training.
  • Lower ATE standard deviation on hard-motion sequences means more predictable trajectory error during aggressive maneuvers, which matters for real navigation.
  • The best curriculum strategy differs by benchmark, so practitioners can choose a scheduler based on the dominant motion patterns of the target deployment.

Reading between the lines

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

  • Equations (5)-(6) apply one scalar $\varphi(L_i)$ to all three weights, so the self-paced variant is mathematically equivalent to an adaptive global loss scale; comparing it against a learning-rate warm-up or decay schedule would show whether the gain is a curriculum effect or optimization dynamics.
  • The RL agents consistently keep flow weight high, which suggests a simpler recipe—training DPVO with a fixed, stronger flow-supervision weight—might reproduce a large share of the improvement without a scheduler.
  • If sample ordering is what matters, other difficulty signals such as blur, illumination, or feature density could replace motion-based scores, extending the method to real-world datasets without ground-truth motion complexity labels.
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 / 4 minor

Summary. The paper proposes three curriculum-learning strategies for training the DPVO monocular visual odometry system: a trajectory-based strategy that orders training data by motion difficulty, a self-paced strategy that modulates loss weights with an exponential function of the current loss, and a DDPG-based adaptive strategy that learns per-component weight schedules. The authors report that their CL-DPVO models improve over the DPVO baseline on TartanAir, EuRoC, TUM-RGBD, and ICL-NUIM, with the self-paced variant achieving state-of-the-art average ATE on the first three benchmarks and the trajectory-based variant leading on ICL-NUIM. They also report faster convergence and reduced overfitting during training.

Significance. If substantiated, the central result would be meaningful: a training-only modification to an existing VO architecture that yields consistent accuracy gains across several benchmarks without changing inference cost. The paper evaluates multiple curriculum variants on four datasets and includes training-dynamics plots for each variant. However, the self-paced equations do not implement the independent per-component curriculum described in the text, and the reported gains are not separated from a global loss-scaling effect. In addition, most benchmark tables lack uncertainty estimates, and no code or trained models are provided, so the headline SOTA claim is not currently established.

major comments (4)
  1. [Section III.D.2, Eqs. (5)-(6)] The self-paced variant does not implement the independent component-wise curriculum claimed in Section III. With w0=0.1 and wF=1 for all components, Eq. (5) gives wf=wp=wr = 0.1 + 0.9 exp(-lambda L_i) at every step. Substituting into Eqs. (3)-(4) yields L_total = w_i (0.1 L_flow + 10 L_translation + 10 w_i L_rotation). The three 'curriculum' weights are therefore identical, and the only effect is a global scalar multiplying the loss, with a squared factor on rotation. The claim in Section IV.C and Table II that this variant outperforms all SOTA methods because of a curriculum over components is unsupported. A control experiment with a scalar loss-scale schedule, or a genuinely independent per-component schedule, is required before this result can be attributed to curriculum learning.
  2. [Section III.D.2 and Section IV.A] Because phi depends only on the current total loss L_i, the mechanism is a loss-dependent global gain, not a per-sample or per-component difficulty curriculum. The text in Section IV.A says the method initially suppresses the impact of difficult samples with high losses, but the weights in Eq. (4) are attached to loss components, not to training samples, and all samples receive the same weights at a given step. To support the self-paced-curriculum interpretation, the authors would need per-sample difficulty weighting as in standard self-paced learning, or independent per-component weights.
  3. [Section IV.A, Figure 2] The trajectory-based comparison is not controlled for compute or data order. The curriculum trains sequentially on easy, medium, and hard subsets and uses early stopping against validation metrics, while the baseline is trained on all data in random order. The reported improvement (ATE 0.17 vs 0.21) could result from the staged data split, longer effective training, or validation-based stopping rather than from curriculum ordering. A baseline trained on the same subsets in random order, or with the same total step budget, is needed.
  4. [Tables II, IV, V, VI] The headline SOTA claim lacks uncertainty quantification. The CL-DPVO rows report median ATE from 5 runs but no per-sequence or cross-run spread, and the differences driving the claim are small (TartanAir average 0.14 m vs 0.17 m; EuRoC average 0.091 m vs 0.105 m). Without confidence intervals, standard deviations, or significance tests, the statement that the method outperforms all other state-of-the-art methods is not established. The authors should report run-to-run variance for the CL models and clarify how baseline and competitor numbers were obtained under identical evaluation settings.
minor comments (4)
  1. [Section II.B and references] The citation '[18]' is used both for Bengio et al.'s 2009 curriculum-learning paper and for Weinshall et al.; in the reference list, [18] is Weinshall et al. and [28] is Bengio et al. Please correct the citation numbering.
  2. [Section IV] The first paragraph of Section IV refers to the 'ECCV 2022 SLAM competition' while Section IV.C refers to the 'ECCV 2020 SLAM competition'; the year should be consistent.
  3. [Table V and Section III.D.1] There are several typographical and grammatical issues, including 'avarage' in Table V, 'While the first approach rely' in Section III.D.1, and the ungrammatical sentence in the Figure 5 caption beginning 'Flow weight maintain high values...'.
  4. [Section IV.A and Figure 6] The claim that the self-paced method achieves a 47% training-time reduction is based on a single training run with no error bars; given the small training-time differences and the lack of repeated runs, this should be stated more cautiously.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the benchmark improvements are externally measured, and the self-paced weights collapsing to a global scalar is a mechanism mismatch rather than a definitional reduction.

full rationale

The paper's central claims are empirical benchmark improvements of DPVO after adding curriculum schedulers. The trajectory-based curriculum assigns difficulty from motion statistics of training sequences and trains sequentially; the test ATE numbers on TartanAir, EuRoC, TUM-RGBD, and ICL-NUIM are measured on disjoint splits and are not encoded in the training loss weights. The self-paced variant defines a progress factor from the current total loss, which makes wf, wp, and wr equal at every step and therefore acts as a global loss-scale schedule rather than an independent per-component curriculum; this is a mechanism and attribution weakness, but it is not circular because the reported ATE improvements are not assumed by Eqs. (5)-(6) and remain external quantities. The DDPG scheduler learns weights from states and rewards and is likewise not circular. There are no load-bearing self-citations: references [5], [6], and [30] are external baselines with no author overlap with the present paper, and no uniqueness theorem or ansatz is imported from the authors' prior work. Hence no derivation step reduces to its own input by construction.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central results depend on several hand-chosen hyperparameters (lambda, thresholds, DDPG settings) and on assumptions that loss magnitude reflects sample difficulty and that the DPVO baseline reproduces correctly, but no new physical entities are introduced.

free parameters (6)
  • Self-paced factor lambda = 0.1
    Empirically tuned in Section III.D.2; controls sensitivity of the exponential progress factor.
  • Initial and final curriculum weights w0, wF = 0.1, 1
    Chosen in Section III.D.2 to bound the interpolation for all weight components.
  • Trajectory difficulty thresholds = 0.44, 0.64
    Selected to divide TartanAir training sequences into equal-sized difficulty groups (Table I).
  • DDPG noise scale = 0.1
    Adaptive exploration noise scaling in Section III.D.3, no sensitivity analysis.
  • DDPG update interval and replay buffer size = k=50, buffer=10k
    Agent training hyperparameters chosen for implementation without ablation.
  • Baseline learning rate adjustment = not specified
    The paper states the learning rate was adjusted for multi-GPU training (Section IV.A), but the value is not reported.
assumptions (6)
  • domain assumption DPVO architecture and default hyperparameters (96 patches, 10-frame window) are taken as given from [5].
    The entire comparison assumes the DPVO baseline is a valid SOTA reference (Section III.B).
  • domain assumption TartanAir ground truth poses and optical flow are accurate enough for supervision.
    Training and evaluation rely on synthetic ground truth from [22] without error analysis.
  • ad hoc to paper Current total loss magnitude Li is a useful measure of learning progress for scheduling weights.
    Section III.D.2 defines phi(Li) = exp(-lambda Li) and treats lower loss as higher progress; no evidence is given that loss magnitude is a reliable difficulty measure.
  • ad hoc to paper The exponential schedule e^{-lambda L} yields a beneficial curriculum.
    Section III.D.2 asserts smoothness and boundedness as advantages; no comparison with other schedules is provided.
  • ad hoc to paper Negative absolute loss is an effective reward for DDPG curriculum agents.
    Section III.D.3 uses ri = -|Li|; no reward-design ablation is given.
  • domain assumption Early stopping on AUC and ATE validation metrics selects the best model fairly for all methods.
    Section IV.A applies early stopping, but the stopping rules may not be identical across baselines.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Robust Monocular Visual Odometry using Curriculum Learning." pith.science (2026). https://pith.science/paper/Y5YZNNHM

@misc{pith2026241113438,
  author       = {Pith},
  title        = {Pith review of: Robust Monocular Visual Odometry using Curriculum Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Y5YZNNHM}},
  note         = {Machine review of arXiv:2411.13438}
}
read the original abstract

Curriculum Learning (CL), drawing inspiration from natural learning patterns observed in humans and animals, employs a systematic approach of gradually introducing increasingly complex training data during model development. Our work applies innovative CL methodologies to address the challenging geometric problem of monocular Visual Odometry (VO) estimation, which is essential for robot navigation in constrained environments. The primary objective of our research is to push the boundaries of current state-of-the-art (SOTA) benchmarks in monocular VO by investigating various curriculum learning strategies. We enhance the end-to-end Deep-Patch-Visual Odometry (DPVO) framework through the integration of novel CL approaches, with the goal of developing more resilient models capable of maintaining high performance across challenging environments and complex motion scenarios. Our research encompasses several distinctive CL strategies. We develop methods to evaluate sample difficulty based on trajectory motion characteristics, implement sophisticated adaptive scheduling through self-paced weighted loss mechanisms, and utilize reinforcement learning agents for dynamic adjustment of training emphasis. Through comprehensive evaluation on the diverse synthetic TartanAir dataset and complex real-world benchmarks such as EuRoC and TUM-RGBD, our Curriculum Learning-based Deep-Patch-Visual Odometry (CL-DPVO) demonstrates superior performance compared to existing SOTA methods, including both feature-based and learning-based VO approaches. The results validate the effectiveness of integrating curriculum learning principles into visual odometry systems.

Figures

Figures reproduced from arXiv: 2411.13438 by the authors.

Figure 1
Figure 1. Distribution of difficulty scores across the TartanAir training dataset, [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Model validation set performance metrics (ATE and AUC) dur [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 4
Figure 4. Model validation set performance metrics (ATE and AUC) during [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figures from the paper (4 more)
Figure 3
Figure 3. Figure 3: Model validation set performance metrics (ATE and AUC) during [PITH_FULL_IMAGE:figures/full_fig_p006_3.png]
Figure 7
Figure 7. Figure 7: The comparison of the Area Under the Curve (AUC) for the validation [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 6
Figure 6. Figure 6: Flow loss progression comparing dynamic curriculum learning [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 8
Figure 8. Figure 8: Visualizing TartanAir test split results for CL-DPVO, DPVO, and [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]

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. AutoLoop: Fast Visual SLAM Fine-tuning through Agentic Curriculum Learning

    cs.RO 2025-01 conditional novelty 4.0 of 10

    AutoLoop uses a DDPG agent to schedule loop-closure loss weights while fine-tuning DPVO on precomputed loop pairs, claiming an order-of-magnitude reduction in training time.

Reference graph

Works this paper leans on

39 extracted references · 25 canonical work pages · cited by 1 Pith paper

  1. [1]

    Posenet: A convolutional network for real-time 6-dof camera relocalization

    Alex Kendall, Matthew Grimes, and Roberto Cipolla. “Posenet: A convolutional network for real-time 6-dof camera relocalization”. In: Proceedings of the IEEE international conference on computer vision . 2015, pp. 2938–2946

    work page 2015

  2. [2]

    Deepvo: Towards end-to-end visual odometry with deep recurrent convolutional neural net- works

    Sen Wang et al. “Deepvo: Towards end-to-end visual odometry with deep recurrent convolutional neural net- works”. In: 2017 IEEE international conference on robotics and automation (ICRA). IEEE. 2017, pp. 2043– 2050

    work page 2017

  3. [3]

    UnDeepVO: Monocular Visual Odom- etry Through Unsupervised Deep Learning

    Ruihao Li et al. “UnDeepVO: Monocular Visual Odom- etry Through Unsupervised Deep Learning”. In: 2018 IEEE International Conference on Robotics and Au- tomation (ICRA). 2018, pp. 7286–7291

    work page 2018

  4. [4]

    Learning monocular visual odometry through geometry-aware curriculum learning

    Muhamad Risqi U Saputra et al. “Learning monocular visual odometry through geometry-aware curriculum learning”. In: 2019 international conference on robotics and automation (ICRA) . IEEE. 2019, pp. 3549–3555

    work page 2019

  5. [5]

    Deep patch visual odometry

    Zachary Teed, Lahav Lipson, and Jia Deng. “Deep patch visual odometry”. In: vol. 36. 2024

    2024

  6. [6]

    Deep event visual odometry

    Simon Klenk et al. “Deep event visual odometry”. In: 2024 International Conference on 3D Vision (3DV) . IEEE. 2024, pp. 739–749

    2024

  7. [7]

    Learning to deblur

    Christian J Schuler et al. “Learning to deblur”. In: IEEE transactions on pattern analysis and machine intelligence 38.7 (2015), pp. 1439–1451

    work page 2015

  8. [8]

    The Effects of Super-Resolution on Object Detection Performance in Satellite Imagery

    Jacob Shermeyer and Adam Van Etten. “The Effects of Super-Resolution on Object Detection Performance in Satellite Imagery”. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). 2019, pp. 1432–1441

    work page 2019

Show all 39 references
  1. [9]

    Quality classified image analysis with application to face detection and recognition

    Fei Yang et al. “Quality classified image analysis with application to face detection and recognition”. In: 2018 24th International Conference on Pattern Recognition (ICPR). IEEE. 2018, pp. 2863–2868

    2018

  2. [10]

    Quality robust mixtures of deep neural networks

    Samuel F Dodge and Lina J Karam. “Quality robust mixtures of deep neural networks”. In: IEEE Trans- actions on Image Processing 27.11 (2018), pp. 5553– 5562

    2018

  3. [11]

    Understanding how image quality affects deep neural networks

    Samuel Dodge and Lina Karam. “Understanding how image quality affects deep neural networks”. In: 2016 eighth international conference on quality of multimedia experience (QoMEX). IEEE. 2016, pp. 1–6

    2016

  4. [12]

    Quality aware network for set to set recognition

    Yu Liu, Junjie Yan, and Wanli Ouyang. “Quality aware network for set to set recognition”. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2017, pp. 5790–5799

    2017

  5. [13]

    Orb-slam3: An accurate open- source library for visual, visual–inertial, and multimap slam

    Carlos Campos et al. “Orb-slam3: An accurate open- source library for visual, visual–inertial, and multimap slam”. In: IEEE Transactions on Robotics 37.6 (2021), pp. 1874–1890

    2021

  6. [14]

    Visual odometry: Part ii: Matching, robustness, optimization, and applications

    Friedrich Fraundorfer and Davide Scaramuzza. “Visual odometry: Part ii: Matching, robustness, optimization, and applications”. In: IEEE Robotics & Automation Magazine 19.2 (2012), pp. 78–90

    2012

  7. [15]

    Event-aided direct sparse odometry

    Javier Hidalgo-Carrió, Guillermo Gallego, and Davide Scaramuzza. “Event-aided direct sparse odometry”. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 2022, pp. 5781–5790

    2022

  8. [16]

    Vinet: Visual-inertial odometry as a sequence-to-sequence learning problem

    Ronald Clark et al. “Vinet: Visual-inertial odometry as a sequence-to-sequence learning problem”. In: Proceed- ings of the AAAI conference on artificial intelligence . V ol. 31. 1. 2017

    2017

  9. [17]

    Robust Visual Odometry Leveraging Mixture of Manhattan Frames in Indoor Environments

    Huayu Yuan et al. “Robust Visual Odometry Leveraging Mixture of Manhattan Frames in Indoor Environments”. In: Sensors 22.22 (2022), p. 8644

    2022

  10. [18]

    Cur- riculum learning by transfer learning: Theory and ex- periments with deep networks

    Daphna Weinshall, Gad Cohen, and Dan Amir. “Cur- riculum learning by transfer learning: Theory and ex- periments with deep networks”. In: International con- ference on machine learning . PMLR. 2018, pp. 5238– 5246

    2018

  11. [19]

    Self-paced curriculum learning

    Lu Jiang et al. “Self-paced curriculum learning”. In: Proceedings of the AAAI Conference on Artificial Intel- ligence. V ol. 29. 1. 2015

    2015

  12. [20]

    Stc: A simple to complex frame- work for weakly-supervised semantic segmentation

    Yunchao Wei et al. “Stc: A simple to complex frame- work for weakly-supervised semantic segmentation”. In: IEEE transactions on pattern analysis and machine intelligence 39.11 (2016), pp. 2314–2320

    2016

  13. [21]

    On the power of curriculum learning in training deep networks

    Guy Hacohen and Daphna Weinshall. “On the power of curriculum learning in training deep networks”. In: International conference on machine learning . PMLR. 2019, pp. 2535–2544

    2019

  14. [22]

    TartanAir: A dataset to push the limits of visual slam

    Wenshan Wang et al. “TartanAir: A dataset to push the limits of visual slam”. In: 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE. 2020, pp. 4909–4916

    2020

  15. [23]

    A benchmark for the evaluation of RGB-D SLAM systems

    Jürgen Sturm et al. “A benchmark for the evaluation of RGB-D SLAM systems”. In: 2012 IEEE/RSJ inter- national conference on intelligent robots and systems . IEEE. 2012, pp. 573–580

    2012

  16. [24]

    The EuRoC micro aerial vehicle datasets

    Michael Burri et al. “The EuRoC micro aerial vehicle datasets”. In: The International Journal of Robotics Research 35.10 (2016), pp. 1157–1163

    2016

  17. [25]

    A benchmark for RGB-D visual odometry, 3D reconstruction and SLAM

    Ankur Handa et al. “A benchmark for RGB-D visual odometry, 3D reconstruction and SLAM”. In: 2014 IEEE international conference on Robotics and automa- tion (ICRA). IEEE. 2014, pp. 1524–1531

    2014

  18. [26]

    Droid-slam: Deep visual slam for monocular, stereo, and rgb-d cameras

    Zachary Teed and Jia Deng. “Droid-slam: Deep visual slam for monocular, stereo, and rgb-d cameras”. In: Advances in neural information processing systems 34 (2021), pp. 16558–16569

    2021

  19. [27]

    Least-squares estimation of trans- formation parameters between two point patterns

    Shinji Umeyama. “Least-squares estimation of trans- formation parameters between two point patterns”. In: IEEE Transactions on Pattern Analysis & Machine Intelligence 13.04 (1991), pp. 376–380

    1991

  20. [28]

    Curriculum learning

    Yoshua Bengio et al. “Curriculum learning”. In: Pro- ceedings of the 26th Annual International Conference on Machine Learning . ICML ’09. Montreal, Quebec, Canada: Association for Computing Machinery, 2009, pp. 41–48

    2009

  21. [29]

    Structure-from-Motion Revisited

    Johannes L. Schönberger and Jan-Michael Frahm. “Structure-from-Motion Revisited”. In: 2016 IEEE 11 Conference on Computer Vision and Pattern Recogni- tion (CVPR). 2016, pp. 4104–4113

    2016

  22. [30]

    Deep Visual Odometry with Events and Frames

    Roberto Pellerito et al. “Deep Visual Odometry with Events and Frames”. In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Sys- tems (IROS. IEEE) . 2024

    2024

  23. [31]

    Pixelwise view se- lection for unstructured multi-view stereo

    Johannes L Schönberger et al. “Pixelwise view se- lection for unstructured multi-view stereo”. In: Com- puter Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part III 14 . Springer. 2016, pp. 501–518

    2016

  24. [32]

    SVO: Fast semi-direct monocular visual odom- etry

    Christian Forster, Matia Pizzoli, and Davide Scara- muzza. “SVO: Fast semi-direct monocular visual odom- etry”. In: 2014 IEEE international conference on robotics and automation (ICRA) . IEEE. 2014, pp. 15– 22

    2014

  25. [33]

    Direct sparse odometry

    Jakob Engel, Vladlen Koltun, and Daniel Cremers. “Direct sparse odometry”. In: IEEE transactions on pattern analysis and machine intelligence 40.3 (2017), pp. 611–625

    2017

  26. [34]

    Tar- tanvo: A generalizable learning-based vo

    Wenshan Wang, Yaoyu Hu, and Sebastian Scherer. “Tar- tanvo: A generalizable learning-based vo”. In: Confer- ence on Robot Learning . PMLR. 2021, pp. 1761–1772

    2021

  27. [35]

    Visual odom- etry

    D. Nister, O. Naroditsky, and J. Bergen. “Visual odom- etry”. In: Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004. V ol. 1. 2004, pp. I–I

    2004

  28. [36]

    Two years of visual odometry on the mars exploration rovers

    Mark Maimone, Yang Cheng, and Larry Matthies. “Two years of visual odometry on the mars exploration rovers”. In: Journal of Field Robotics 24.3 (2007), pp. 169–186

    2007

  29. [37]

    Visual odometry by multi-frame feature integration

    Hernán Badino, Akihiro Yamamoto, and Takeo Kanade. “Visual odometry by multi-frame feature integration”. In: Proceedings of the IEEE International Conference on Computer Vision Workshops . 2013, pp. 222–229

    2013

  30. [38]

    Self-paced learning with diversity

    Lu Jiang et al. “Self-paced learning with diversity”. In: Advances in neural information processing systems 27 (2014)

    2014

  31. [39]

    A general optimization-based frame- work for global pose estimation with multiple sensors

    Tong Qin et al. “A general optimization-based frame- work for global pose estimation with multiple sensors”. In: arXiv preprint arXiv:1901.03642 (2019)

    2019 arXiv

Pith tools

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