Pith. sign in

REVIEW 3 major objections 2 minor 53 references

Simulation-based inference using splitting schemes for partially observed diffusions in chemical reaction networks

T0 review · 3 major / 2 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Chemical Langevin equations can be rewritten as perturbed CIR-type SDEs, enabling a structure-preserving splitting scheme that makes Bayesian inference for partially observed reaction networks computationally cheaper.

desk verdict The abstract describes a plausible and useful method, but we can only judge the abstract; the invariant-distribution claim is the key thing to check. read the letter →

arxiv 2508.11438 v1 pith:IZTA6NBC submitted 2025-08-15 stat.ME stat.CO

Petar Jovanovski , Andrew Golightly , Umberto Picchini , Massimiliano Tamborrino This is my paper
classification stat.MEstat.CO MSC 62F1560H3565C30
keywords approximateBayesiancomputationABC-SMCchemicalLangevinequationCox-Ingersoll-Rossprocesssplittingschemepartiallyobserveddiffusionssummarystatisticsreactionnetworks
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 tackles two problems in chemical reaction networks: simulating the chemical Langevin equation (an SDE with multiplicative, non-commutative noise) and doing parameter inference when only some species are observed, with measurement error. It shows each coordinate of the CLE, conditioned on the rest, is a Cox-Ingersoll-Ross-like SDE with linear drift and square-root diffusion, plus extra Brownian perturbations. On this class it builds a numerical splitting scheme that preserves state space, oscillations, and invariant distributions even at large integration steps, unlike Euler-Maruyama. It wraps this in an ABC-SMC algorithm with data-conditional simulation and sequentially learned summary statistics, allowing inference for multidimensional partially observed systems at reduced cost. The results on Repressilator, Lotka-Volterra, and two-pool models suggest this could make Bayesian inference feasible for realistic reaction networks.

What carries the argument

The key machinery is the splitting scheme for perturbed conditionally CIR-type SDEs: each coordinate is simulated as a CIR process with linear drift and square-root diffusion, with the cross-coordinate coupling treated as additive Brownian perturbations handled in separate substeps. This preserves the state space (nonnegativity) and the invariant distribution, allowing large integration steps. The other pillar is the ABC-SMC algorithm with data-conditional simulation and sequential learning of summary statistics, which makes the method scale to partially observed systems.

What would settle it

Run the splitting scheme with a large step on a two-species reaction network with known stationary distribution, and compare the empirical invariant measure against a finely integrated Euler-Maruyama or exact Gillespie simulation; if the invariant distribution drifts with step size well before the ABC tolerance, the central claim fails.

Watch

Extended reading notes

Core claim

The central discovery is a structural rewriting: for the chemical Langevin equation, conditioning each coordinate on the others yields a 'perturbed conditionally CIR-type' SDE—linear drift, square-root diffusion, plus extra Brownian terms. The paper develops a splitting scheme that treats these components separately, preserving qualitative features such as the state space, oscillatory behavior, and invariant distribution at step sizes where Euler-Maruyama fails. Combined with an ABC-SMC sampler that simulates from the data-conditional process and learns summary statistics adaptively, this gives a practical inference method for partially observed multidimensional reaction networks. The paper'

Load-bearing premise

The method relies on the splitting scheme preserving the true invariant distribution of the CLE at the large integration steps used by the ABC sampler, and on the adaptively learned summary statistics carrying enough information about the parameters.

Editorial extensions

If this is right

  • Bayesian inference for the stochastic Repressilator and similar oscillatory gene networks becomes feasible with fewer simulation steps.
  • The splitting scheme can replace Euler-Maruyama in ABC pipelines, reducing computational cost while maintaining posterior accuracy.
  • The method extends approximate Bayesian computation to multidimensional partially observed reaction networks without hand-crafted summary statistics.
  • Large integration steps, previously causing instability, can be used safely for these CLEs.

Reading between the lines

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

  • The conditional CIR structure may extend beyond chemical reaction networks to other SDEs with similar drift/diffusion forms, offering a general simulation strategy.
  • If the invariant distribution preservation holds at very large steps, the scheme could be used for steady-state sampling without fine time-stepping.
  • The sequential summary statistic learning could be combined with other simulation schemes, possibly improving their data efficiency.
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

3 major / 2 minor

Summary. The paper claims three contributions for simulation and inference of chemical reaction networks described by the chemical Langevin equation: (1) rewriting such SDEs as perturbed conditionally Cox-Ingersoll-Ross-type SDEs; (2) developing a numerical splitting scheme that preserves structural properties, including oscillations, state space, and invariant distributions, and is robust for large integration time steps; and (3) proposing an ABC-SMC algorithm with data-conditional simulation and sequential learning of summary statistics for partially observed multidimensional systems. Validation is reported on three models: stochastic Repressilator, Lotka-Volterra, and two-pool systems. However, the full text supplied with the submission is not the paper under review: it is an unrelated indoor-navigation manuscript. Consequently, only the abstract is available for assessment, and the derivations, simulation studies, and error analyses cannot be inspected.

Significance. If the claims hold, the paper would address a real bottleneck in simulation-based inference for reaction networks: Euler-Maruyama schemes typically require very small time steps for CIR-type SDEs, and reliable inference for partially observed multidimensional diffusion models remains hard. A splitting scheme that provably preserves invariant distributions at large steps would be a meaningful contribution, and an ABC-SMC sampler using learned summaries could extend Bayesian inference to realistic partially observed settings. That said, the paper's significance cannot be evaluated from the abstract alone; the structural-preservation claim and the ABC posterior validity are nontrivial and need full derivations and experiments. No machine-checked proofs, code, or reproducibility artifacts are visible in the provided material.

major comments (3)
  1. [Full text (provided as arXiv:2508.11446)] The submitted full text is not the manuscript described by the abstract. It is a computer-vision paper on indoor navigation titled "Inside Knowledge: Graph-based Path Generation with Explainable Data Augmentation and Curriculum Learning for Visual Indoor Navigation." The splitting scheme, ABC-SMC algorithm, and the three validation studies are therefore unavailable. This is an editorial obstruction that prevents technical review; the correct manuscript must be supplied before any soundness assessment can be made.
  2. [Abstract, second contribution] The central numerical claim is that the splitting scheme "preserves structural properties of the model, such as oscillations, state space and invariant distributions" and is "robust for large integration time steps." No proof, theorem, or error bound is provided in the abstract. For CIR-type SDEs with additional Brownian perturbations, invariant-measure preservation under a splitting scheme is not automatic; it can depend on step size through boundary behavior at zero and through coordinate coupling in the split composition. Without the full derivation, it is not possible to verify that the discrete scheme's invariant measure coincides with the CLE's invariant measure. This is load-bearing because the ABC sampler uses the simulated trajectories as a model proxy; if the invariant distribution is biased, the posterior targets the wrong process.
  3. [Abstract, third contribution] The ABC-SMC algorithm relies on "sequential learning of summary statistics." In ABC, the choice and informativeness of summary statistics determine the quality of the approximate posterior. The abstract does not specify how summaries are learned, what guarantees they satisfy, or how "data-conditional simulation" handles partial observations in practice. The three example models (Repressilator, Lotka-Volterra, two-pool) are finite demonstrations; they do not establish that posterior inference is unbiased at the selected integration step sizes or ABC tolerances. Quantitative comparison against ground-truth posteriors and a step-size-bias analysis are needed to support the inferential accuracy claim.
minor comments (2)
  1. [Abstract, terminology] The term "perturbed conditionally Cox-Ingersoll-Ross-type SDEs" is used without a formal definition. Since the entire first contribution depends on this class, a precise definition and an example for the chemical Langevin equation would improve clarity.
  2. [Abstract, performance claim] The abstract claims "reduced computational cost" but gives no quantitative basis. A comparison of wall-clock times or complexity with Euler-Maruyama-based ABC would make the claim verifiable.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the abstract's CLE-to-splitting-scheme-to-ABC derivation is forward-looking and externally validated, with no fitted-parameter predictions or load-bearing self-citations identified.

full rationale

The abstract's claimed derivation chain — rewriting the chemical Langevin equation as a perturbed conditionally CIR-type SDE, designing a splitting scheme that preserves state space/invariant distributions, and embedding this in an ABC-SMC sampler — is presented as a constructive mathematical and computational contribution, validated on synthetic data from the Repressilator, Lotka-Volterra, and two-pool models. I find no step in the abstract where an input is defined in terms of an output, no fitted parameter that is later relabeled as a prediction, and no load-bearing self-citation or imported uniqueness theorem. The validation against known model dynamics and synthetic data provides external benchmarks rather than circular confirmation. Note that the full text supplied with the query is a different manuscript (a computer-vision indoor-navigation paper), so the equations and proofs behind the abstract could not be inspected; however, missing supporting text is a completeness/verification issue, not evidence of circularity. Under the hard rules, no circular step can be exhibited, so the score is 0.

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

No free parameters are visible from the abstract. The ABC algorithm estimates model parameters rather than introducing ad hoc fitted constants. Algorithmic tuning parameters (tolerance, number of particles, summary statistic dimension) are not specified in the abstract. The abstract introduces no new entities such as particles, mediators, or conserved quantities. The 'perturbed CIR-type' representation is a reformulation, not a new entity.

assumptions (3)
  • domain assumption The chemical Langevin equation is a valid SDE representation of the chemical reaction network dynamics.
    The paper's simulation and inference all operate on the CLE; if the CLE is not a faithful model of the true reaction network, the results do not transfer to the physical system. This is a standard assumption in stochastic kinetics but is stated nowhere in the abstract.
  • standard math The numerical splitting scheme converges to the true CLE solution as the time step goes to zero, and its invariant distribution matches the true invariant distribution at the step sizes used.
    Any numerical approximation of an SDE relies on consistency and convergence properties. The paper claims preservation of invariant distributions at large steps, which is a strong property that must hold for the ABC inference to be unbiased. The abstract does not provide proof.
  • domain assumption The sequentially learned summary statistics and the ABC tolerance produce posterior approximations close to the true posterior.
    ABC relies on summary statistics being sufficiently informative and on a small enough tolerance. The abstract says summaries are learned sequentially but does not state sufficiency guarantees. If summaries are not sufficient, inferred posteriors can be misleading.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Simulation-based inference using splitting schemes for partially observed diffusions in chemical reaction networks." pith.science (2026). https://pith.science/paper/IZTA6NBC

@misc{pith2026250811438,
  author       = {Pith},
  title        = {Pith review of: Simulation-based inference using splitting schemes for partially observed diffusions in chemical reaction networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IZTA6NBC}},
  note         = {Machine review of arXiv:2508.11438}
}
read the original abstract

We address the problem of simulation and parameter inference for chemical reaction networks described by the chemical Langevin equation, a stochastic differential equation (SDE) representation of the dynamics of the chemical species. This is challenging for two main reasons. First, the (multi-dimensional) SDEs cannot be explicitly solved and are driven by multiplicative and non-commutative noise, requiring the development of advanced numerical schemes for their approximation and simulation. Second, not all components of the SDEs are directly observed, as the available discrete-time data are typically incomplete and/or perturbed with measurement error. We tackle these issues via three contributions. First, we show that these models can be rewritten as perturbed conditionally Cox-Ingersoll-Ross-type SDEs, i.e., each coordinate, conditioned on all other coordinates being fixed, follows an SDE with linear drift and square root diffusion coefficient perturbed by additional Brownian motions. Second, for this class of SDEs, we develop a numerical splitting scheme that preserves structural properties of the model, such as oscillations, state space and invariant distributions, unlike the commonly used Euler-Maruyama scheme. Our numerical method is robust for large integration time steps. Third, we propose a sequential Monte Carlo approximate Bayesian computation algorithm incorporating "data-conditional" simulation and sequential learning of summary statistics, allowing inference for multidimensional partially observed systems, further developing previous results on fully observed systems based on the Euler-Maruyama scheme. We validate our approach on models of interest in chemical reaction networks, such as the stochastic Repressilator, Lotka-Volterra, and two-pool systems, demonstrating its effectiveness, in terms of both numerical and inferential accuracy, and reduced computational cost.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

53 extracted references · 48 canonical work pages

  1. [1]

    A. A. Alajami, G. Moreno, and R. Pous. Design of a uav for autonomous rfid-based dynamic inventories using stigmergy for mapless indoor environments. Drones, 6(8), 2022. 2

    work page 2022

  2. [2]

    E. J. Alqahtani, F. H. Alshamrani, H. F. Syed, and F. A. Al- haidari. Survey on algorithms and techniques for indoor nav- igation systems. In 2018 21st Saudi Computer Society Na- tional Computer Conference (NCC), pages 1–9, 2018. 2

    work page 2018

  3. [3]

    Amini, G

    A. Amini, G. Rosman, S. Karaman, and D. Rus. Varia- tional end-to-end navigation and localization. In 2019 In- ternational Conference on Robotics and Automation (ICRA), pages 8958–8964, 2019. 2

    work page 2019

  4. [4]

    B. Ando, S. Baglio, V . Marletta, and N. Pitrone. A mixed in- ertial & rfid orientation tool for the visually impaired. In Proceedings of the 6th International Multi-Conference on Systems, Signals and Devices, Djerba, Tunisia, March 2009. Held on 23–26 March 2009. 2

    work page 2009

  5. [5]

    S. Anup, A. Goel, and S. Padmanabhan. Visual positioning system for automated indoor/outdoor navigation. In TEN- CON 2017 - 2017 IEEE Region 10 Conference, pages 1027– 1031, 2017. 2

    work page 2017

  6. [6]

    M. Y . Arafat, M. M. Alam, and S. Moh. Vision-based navi- gation techniques for unmanned aerial vehicles: Review and challenges. Drones, 7(2), 2023. 2

    work page 2023

  7. [7]

    Barua, C

    B. Barua, C. Gomes, S. Baghe, and J. Sisodia. A self-driving car implementation using computer vision for detection and navigation. In 2019 International Conference on Intelli- gent Computing and Control Systems (ICCS) , pages 271– 274, 2019. 2

    work page 2019

  8. [8]

    Beydoun, V

    K. Beydoun, V . Felea, and H. Guyennet. Wireless sensor network system helping navigation of the visually impaired. In Proceedings of the IEEE International Conference on In- formation and Communication Technologies: From Theory to Applications, Damascus, Syria, April 2008. Held on 7–11 April 2008. 2

    work page 2008

Show all 53 references
  1. [9]

    Botta and G

    A. Botta and G. Quaglia. Performance analysis of low-cost tracking system for mobile robots. Machines, 8(2):29, 2020. 2

    2020

  2. [10]

    Chang, S

    Y . Chang, S. Tsai, and Y . Wang. A context aware hand- held wayfinding system for individuals with cognitive im- pairments. In Proceedings of the 10th International ACM SIGACCESS Conference on Computers and Accessibility , Halifax, NS, Canada, October 2008. Held on 13–15 Octo...

    2008

  3. [11]

    P. Corke. Robotics, Vision and Control: Fundamental Al- gorithms in MATLAB, volume 73 of Springer Tracts in Ad- vanced Robotics. Springer, 2011. 3

    2011

  4. [12]

    Corrales-Paredes, M

    A. Corrales-Paredes, M. Malfaz, and M. Salichs. Signage system for the navigation of autonomous robots in indoor environments. Industrial Informatics, IEEE Transactions on, 10:680–688, 02 2014. 2

    2014

  5. [13]

    Devries and G

    T. Devries and G. W. Taylor. Improved regularization of convolutional neural networks with cutout. CoRR, abs/1708.04552, 2017. 6

    2017 arXiv

  6. [14]

    Do, M.-T

    T.-D. Do, M.-T. Duong, Q.-V . Dang, and M.-H. Le. Real- time self-driving car navigation using deep neural network. In 2018 4th International Conference on Green Technology and Sustainable Development (GTSD), pages 7–12, 2018. 2

    2018

  7. [15]

    Dosovitskiy, P

    A. Dosovitskiy, P. Fischer, E. Ilg, P. Hausser, C. Hazirbas, V . Golkov, P. Van Der Smagt, D. Cremers, and T. Brox. Flownet: Learning optical flow with convolutional networks. In Proceedings of the IEEE international conference on com- puter vision, pages 2758–2766, 2015. 4

    2015

  8. [16]

    El-Sheimy and Y

    N. El-Sheimy and Y . Li. Indoor navigation: state of the art and future trends. Satellite Navigation, 2(7), 2021. 2

    2021

  9. [17]

    Endres, J

    F. Endres, J. Hess, N. Engelhard, J. Sturm, D. Cremers, and W. Burgard. An evaluation of the rgb-d slam system. In2012 IEEE International Conference on Robotics and Automation, pages 1691–1696, 2012. 2

    2012

  10. [18]

    Fusco, S

    G. Fusco, S. A. Cheraghi, L. Neat, and J. M. Coughlan. An indoor navigation app using computer vision and sign recognition. In Computers Helping People with Special Needs, volume 12376 ofLecture Notes in Computer Science, pages 485–494, September 2020. Epub 2020 Sep 4. PMID: 3...

    2020

  11. [19]

    T. Guo, J. Dong, H. Li, and Y . Gao. Simple convolutional neural network on image classification. In 2017 IEEE 2nd International Conference on Big Data Analysis (ICBDA) , pages 721–724, 2017. 5

    2017

  12. [20]

    I. A. Hashish, G. Motta, M. Meazza, G. Bu, K. Liu, L. Duico, and A. Longo. Navapp: An indoor navigation application: A smartphone application for libraries. In 2017 14th Workshop on Positioning, Navigation and Communications (WPNC) , pages 1–6, 2017. 1

    2017

  13. [21]

    Huang and G

    H. Huang and G. Gartner. A Survey of Mobile Indoor Navi- gation Systems, pages 305–319. Springer Berlin Heidelberg, Berlin, Heidelberg, 2010. 2

    2010

  14. [22]

    Q. Jin, Y . Liu, Y . Man, and F. Li. Visual slam with rgb-d cameras. In 2019 Chinese Control Conference (CCC), pages 4072–4077, 2019. 2

    2019

  15. [23]

    N. Kanwal. A navigation system for visually impaired: A fu- sion of vision and depth sensor. Applied Bionics and Biome- chanics, 2015, 08 2015. 2

    2015

  16. [24]

    A. G. Karkar, S. Al-Maadeed, J. Kunhoth, et al. Camnav: A computer-vision indoor navigation system. Journal of Su- percomputing, 77:7737–7756, 2021. 2

    2021

  17. [25]

    D. Khan, Z. Cheng, H. Uchiyama, S. Ali, M. Asshad, and K. Kiyokawa. Recent advances in vision-based indoor nav- igation: A systematic literature review. Computers and Graphics, 104:24–45, 2022. 2 9

    2022

  18. [26]

    D. Khan, S. Ullah, and S. Nabi. A generic approach toward indoor navigation and pathfinding with robust marker track- ing. Remote Sensing, 11(24):3052, 2019. 2

    2019

  19. [27]

    Kjærgaard, H

    M. Kjærgaard, H. Blunck, T. Godsk, T. Toftkjær, D. Lund, and K. Grønbæk. Indoor positioning using gps revisited. pages 38–56, 05 2010. 1

    2010

  20. [28]

    Kumar, T

    A. Kumar, T. Saini, P. B. Pandey, et al. Vision-based out- door navigation of self-driving car using lane detection. In- ternational Journal of Information Technology, 14:215–227,

  21. [29]

    Y . H. Lee and G. Medioni. Rgb-d camera based wearable navigation system for the visually impaired. Computer Vi- sion and Image Understanding, 149:3–20, 2016. Special is- sue on Assistive Computer Vision and Robotics - ”Assistive Solutions for Mobility, Communication and HMI”. 2

    2016

  22. [30]

    Leordeanu and I

    M. Leordeanu and I. Paraicu. Driven by vision: Learning navigation by visual localization and trajectory prediction. Sensors, 21(3), 2021. 2

    2021

  23. [31]

    B. Li, J. P. Mu ˜noz, X. Rong, Q. Chen, J. Xiao, Y . Tian, A. Arditi, and M. Yousuf. Vision-based mobile indoor as- sistive navigation aid for blind people. IEEE Transactions on Mobile Computing, 18(3):702–714, 2019. 2

    2019

  24. [32]

    F. Li, C. Guo, B. Luo, and H. Zhang. Multi goals and multi scenes visual mapless navigation in indoor using meta- learning and scene priors. Neurocomputing, 449:368–377,

  25. [33]

    J. M. Loomis, R. G. Golledge, R. L. Klatzky, and J. R. Marston. Assisting wayfinding in visually impaired travel- ers. In Applied Spatial Cognition: From Research to Cog- nitive Technology, pages 179–202. Lawrence Erlbaum Asso- ciates, Mahwah, NJ, USA, 2006. 2

    2006

  26. [34]

    L ´opez, P

    J. L ´opez, P. S´anchez-Vilari˜no, R. Sanz, and E. Paz. Efficient local navigation approach for autonomous driving vehicles. IEEE Access, 9:79776–79792, 2021. 2

    2021

  27. [35]

    Manduchi, S

    R. Manduchi, S. Kurniawan, and H. Bagherinia. Blind guidance using mobile computer vision: A usability study. In Proceedings of the 12th International ACM SIGACCESS Conference on Computers and Accessibility , Orlando, FL, USA, October 2010. Held on 25–27 October 2010. 2

    2010

  28. [36]

    M. L. Mekhalfi, F. Melgani, A. Zeggada, F. G. De Natale, M. A.-M. Salem, and A. Khamis. Recovering the sight to blind people in indoor environments with smart technolo- gies. Expert Systems with Applications , 46:129–138, 2016. 2

    2016

  29. [37]

    Morar, A

    A. Morar, A. Moldoveanu, I. Mocanu, F. Moldoveanu, I. E. Radoi, V . Asavei, A. Gradinaru, and A. Butean. A com- prehensive survey of indoor localization methods based on computer vision. Sensors, 20(9), 2020. 2

    2020

  30. [38]

    R. P. Padhy, S. Verma, S. Ahmad, S. K. Choudhury, and P. K. Sa. Deep neural network for autonomous uav navigation in indoor corridor environments. Procedia computer science , 133:643–650, 2018. 2

    2018

  31. [39]

    Redmon, S

    J. Redmon, S. K. Divvala, R. B. Girshick, and A. Farhadi. You only look once: Unified, real-time object detection. CoRR, abs/1506.02640, 2015. 6

    2015 arXiv

  32. [40]

    Royer, M

    E. Royer, M. Lhuillier, M. Dhome, et al. Monocular vision for mobile robot localization and autonomous navigation. International Journal of Computer Vision , 74(3):237–260,

  33. [41]

    Ruotsalainen

    L. Ruotsalainen. Vision-Aided Pedestrian Naviga- tion for Challenging GNSS Environments . Suomen geodeettisen laitoksen julkaisuja - Publications of the Finnish Geodetic Institute;151. Suomen geodeettinen laitos, Nov. 2013. Awarding institution:Tampereen teknillinen yliopisto ...

    2013

  34. [42]

    R. R. Selvaraju, A. Das, R. Vedantam, M. Cogswell, D. Parikh, and D. Batra. Grad-cam: Why did you say that? visual explanations from deep networks via gradient-based localization. CoRR, abs/1610.02391, 2016. 6

    2016 arXiv

  35. [43]

    Soviany, R

    P. Soviany, R. T. Ionescu, P. Rota, and N. Sebe. Curriculum learning: A survey. CoRR, abs/2101.10382, 2021. 6

    2021 arXiv

  36. [44]

    J. T. Springenberg, A. Dosovitskiy, T. Brox, and M. Ried- miller. Striving for simplicity: The all convolutional net,

  37. [45]

    X. Teng, D. Guo, Y . Guo, X. Zhou, and Z. Liu. Cloudnavi: Toward ubiquitous indoor navigation service with 3d point clouds. ACM Trans. Sen. Netw., 15(1), Jan. 2019. 2

    2019

  38. [46]

    TREUILLET and E

    S. TREUILLET and E. ROYER. Outdoor/indoor vision- based localization for blind pedestrian navigation assistance. International Journal of Image and Graphics , 10(04):481– 496, 2010. 2

    2010

  39. [47]

    Viola, K

    M. Viola, K. Qu, N. Metzger, B. Ke, A. Becker, K. Schindler, and A. Obukhov. Marigold-dc: Zero-shot monocular depth completion with guided diffusion. arXiv preprint arXiv:2412.13389, 2024. 4

    2024 arXiv

  40. [48]

    A. Vu, A. Ramanandan, A. Chen, J. A. Farrell, and M. Barth. Real-time computer vision/dgps-aided inertial navigation system for lane-level vehicle navigation. IEEE Transactions on Intelligent Transportation Systems, 13(2):899–913, 2012. 2

    2012

  41. [49]

    A. Zeb, S. Ullah, and I. Rabbi. Indoor vision-based audi- tory assistance for blind people in semi-controlled environ- ments. In Proceedings of the 4th International Conference on Image Processing Theory, Tools and Applications, Paris, France, October 2014. Held on 14–17 Octob...

    2014

  42. [50]

    Zhang, W

    J. Zhang, W. Liu, and Y . Wu. Novel technique for vision- based uav navigation. IEEE Transactions on Aerospace and Electronic Systems, 47(4):2731–2741, 2011. 2

    2011

  43. [51]

    Zheng, G

    Y . Zheng, G. Shen, L. Li, C. Zhao, M. Li, and F. Zhao. Travi- navi: Self-deployable indoor navigation system. IEEE/ACM Transactions on Networking, 25(5):2655–2669, 2017. 2 10

    2017

  44. [2007]

    Received: 14 October 2005; Accepted: 11 December 2006; Published: 13 January 2007; Issue Date: September

    2005

  45. [2022]

    Received: 05 December 2020; Accepted: 15 July 2021; Published: 04 August 2021; Issue Date: February

    2020

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.