pith. sign in

arxiv: 2604.24010 · v1 · submitted 2026-04-27 · 📊 stat.ME · eess.SP

Efficient Implementations of Extended Object PMBM Filters with Blocked Gibbs Sampling

Yuxuan Xia , \'Angel F. Garc\'ia-Fern\'andez , Lennart Svensson This is my paper

Pith reviewed 2026-05-08 02:23 UTC · model grok-4.3

classification 📊 stat.ME eess.SP
keywords objectextendedpmbmgibbspoissonsamplingblockedvariables
0
0 comments X

The pith

Blocked Gibbs sampling and a collapsed variant provide efficient implementations of the extended object PMBM filter update, matching particle belief propagation performance at lower runtime under the gamma Gaussian inverse-Wishart model.

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

Multiple object tracking from sensors like radar or cameras involves figuring out which measurements belong to which objects and handling new or disappearing objects. The Poisson multi-Bernoulli mixture filter offers an exact Bayesian way to do this for certain models, but the data association step is computationally hard. The authors reformulate the problem using auxiliary variables and the Poisson measurement model to create a joint distribution that factors nicely. They then use blocked Gibbs sampling to draw high-probability association hypotheses quickly. A collapsed version marginalizes out object existence variables for even faster sampling of new objects. Simulations show these methods track objects as well as particle-based alternatives but run much faster. The work focuses on practical implementation under a common extended object model.

Core claim

By formulating the PMBM density on an augmented state space with auxiliary variables and leveraging the Poisson object measurement model, we first derive a joint posterior over potential objects, previous global hypotheses, and current measurement association variables, together with its corresponding factorization. This factorized representation leads to blocked Gibbs samplers that efficiently generate high-weight global hypotheses and thereby provide an efficient implementation of the PMBM update step.

Load-bearing premise

The Poisson object measurement model holds and the gamma Gaussian inverse-Wishart representation accurately captures the extended object states, allowing the derived factorization and Gibbs samplers to produce unbiased high-weight hypotheses without convergence issues in practical scenarios.

Figures

Figures reproduced from arXiv: 2604.24010 by \'Angel F. Garc\'ia-Fern\'andez, Lennart Svensson, Yuxuan Xia.

Figure 1
Figure 1. Figure 1: Diagram of the extended object PMBM filter based on view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of the ground-truth scenario. Two objects are view at source ↗
Figure 3
Figure 3. Figure 3: Performance comparison of PMBM-C-Gibbs2 (with 500 sampling iterations) and PMB-BP (with 10000 particles), showing the GOSPA view at source ↗
read the original abstract

This paper considers multiple extended object tracking based on Poisson multi-Bernoulli mixture (PMBM) filtering, which gives the closed-form Bayesian solution for standard multiple extended object models with Poisson birth. To efficiently address the challenging extended object data association problem in PMBM filtering, we develop implementations of the extended object PMBM filter using blocked Gibbs sampling. By formulating the PMBM density on an augmented state space with auxiliary variables and leveraging the Poisson object measurement model, we first derive a joint posterior over potential objects, previous global hypotheses, and current measurement association variables, together with its corresponding factorization. This factorized representation leads to blocked Gibbs samplers that efficiently generate high-weight global hypotheses and thereby provide an efficient implementation of the PMBM update step. We further introduce a collapsed Gibbs sampling variant, in which the Bernoulli object existence variables are marginalized out, yielding higher sampling efficiency, especially for the initiation of newly detected objects. The proposed methods, implemented under the gamma Gaussian inverse-Wishart model, are compared with an extended object Poisson multi-Bernoulli filter based on particle belief propagation. Simulation results demonstrate that the proposed approaches achieve comparable tracking performance while requiring substantially less runtime.

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

1 major / 0 minor

Summary. The paper proposes efficient blocked Gibbs sampling implementations for the extended object PMBM filter. By augmenting the PMBM density with auxiliary variables and exploiting the Poisson object measurement model, it derives a factorized joint posterior over potential objects, prior global hypotheses, and current measurement associations. This factorization yields blocked Gibbs samplers (plus a collapsed variant that marginalizes Bernoulli existence variables) for the PMBM update. Implemented under the gamma-Gaussian-inverse-Wishart model, the methods are shown in simulations to achieve tracking performance comparable to an extended-object Poisson multi-Bernoulli filter using particle belief propagation, while requiring substantially less runtime.

Significance. If the samplers reliably produce high-weight hypotheses at practical iteration counts, the work supplies a principled, conjugacy-exploiting alternative to particle-based data association in extended-object PMBM filtering. The explicit derivation of the joint posterior and the collapsed-Gibbs variant are clear technical strengths that could improve computational scalability for multi-object tracking with Poisson birth and extended-object measurements.

major comments (1)
  1. [Simulation results] The central efficiency claim rests on the blocked Gibbs samplers generating high-weight global hypotheses in practical runtimes, yet the simulation results report only aggregate runtime gains versus particle BP without any mixing diagnostics (autocorrelation, effective sample size, burn-in length, or trace plots) for the discrete association blocks. In high-dimensional multimodal association spaces this omission leaves the practical usability of the factorization unverified.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback on our manuscript. We address the major comment below and will revise the paper accordingly to strengthen the presentation of the simulation results.

read point-by-point responses

  1. Referee: [Simulation results] The central efficiency claim rests on the blocked Gibbs samplers generating high-weight global hypotheses in practical runtimes, yet the simulation results report only aggregate runtime gains versus particle BP without any mixing diagnostics (autocorrelation, effective sample size, burn-in length, or trace plots) for the discrete association blocks. In high-dimensional multimodal association spaces this omission leaves the practical usability of the factorization unverified.

    Authors: We acknowledge that the current simulations report only aggregate tracking performance and runtime comparisons without explicit mixing diagnostics for the discrete association variables. While the observed equivalence in tracking accuracy to particle belief propagation together with the lower runtimes provides indirect support that the blocked Gibbs samplers produce useful high-weight hypotheses, we agree that direct diagnostics would more rigorously verify sampler behavior in high-dimensional multimodal spaces. In the revised manuscript we will add autocorrelation functions, effective sample sizes, burn-in lengths, and representative trace plots for the association blocks in the simulation section. revision: yes

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard domain assumptions from multi-object tracking literature rather than new postulates; no free parameters or invented entities are introduced in the abstract.

axioms (2)
  • domain assumption The Poisson multi-Bernoulli mixture density gives the closed-form Bayesian solution for standard multiple extended object models with Poisson birth.
    Directly stated in the abstract as the foundation for the filtering approach.
  • domain assumption The gamma Gaussian inverse-Wishart model is a suitable representation for extended objects.
    Used for the proposed implementations and simulations.

pith-pipeline@v0.9.0 · 5515 in / 1347 out tokens · 51916 ms · 2026-05-08T02:23:20.688915+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

62 extracted references · 62 canonical work pages

  1. [1]

    Bar-Shalom, P

    Y . Bar-Shalom, P. K. Willett, and X. Tian,Tracking and data fusion. YBS publishing Storrs, CT, USA:, 2011, vol. 11

    work page 2011

  2. [2]

    Message passing algorithms for scalable multitarget tracking,

    F. Meyer, T. Kropfreiter, J. L. Williams, R. Lau, F. Hlawatsch, P. Braca, and M. Z. Win, “Message passing algorithms for scalable multitarget tracking,”Proceedings of the IEEE, vol. 106, no. 2, pp. 221–259, 2018

    work page 2018

  3. [3]

    R. L. Streit, R. B. Angle, and M. Efe,Analytic combinatorics in multiple object tracking. Springer, 2021

    work page 2021

  4. [4]

    De- tection and tracking meet drones challenge,

    P. Zhu, L. Wen, D. Du, X. Bian, H. Fan, Q. Hu, and H. Ling, “De- tection and tracking meet drones challenge,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 11, pp. 7380–7399, 2021

    work page 2021

  5. [5]

    Soccernet-tracking: Multiple object tracking dataset and benchmark in soccer videos,

    A. Cioppa, S. Giancola, A. Deliege, L. Kang, X. Zhou, Z. Cheng, B. Ghanem, and M. Van Droogenbroeck, “Soccernet-tracking: Multiple object tracking dataset and benchmark in soccer videos,” inProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 3491–3502

    work page 2022

  6. [6]

    Vehicle-to-everything cooperative perception for autonomous driving,

    T. Huang, J. Liu, X. Zhou, D. C. Nguyen, M. R. Azghadi, Y . Xia, Q.-L. Han, and S. Sun, “Vehicle-to-everything cooperative perception for autonomous driving,”Proceedings of the IEEE, vol. 113, no. 5, pp. 443–477, 2025

    work page 2025

  7. [7]

    Extended object tracking: Introduction, overview and applications,

    K. Granstr ¨om, M. Baum, and S. Reuter, “Extended object tracking: Introduction, overview and applications,”Journal of Advances in Information Fusion, vol. 12, no. 2, 2017

    work page 2017

  8. [8]

    A tutorial on multiple extended object tracking,

    K. Granstr ¨om and M. Baum, “A tutorial on multiple extended object tracking,”TechRxiv, 2022

    work page 2022

  9. [9]

    Spatial distribution model for tracking extended objects,

    K. Gilholm and D. Salmond, “Spatial distribution model for tracking extended objects,”IEE Proceedings-Radar , Sonar and Navigation, vol. 152, no. 5, pp. 364–371, 2005

    work page 2005

  10. [10]

    Multiple-hypothesis tracking for targets producing multiple measurements,

    S. P. Coraluppi and C. A. Carthel, “Multiple-hypothesis tracking for targets producing multiple measurements,”IEEE Transactions on Aerospace and Electronic Systems, vol. 54, no. 3, pp. 1485– 1498, 2018

    work page 2018

  11. [11]

    Joint probabilistic data association tracker for extended target tracking applied to X-band marine radar data,

    G. Vivone and P. Braca, “Joint probabilistic data association tracker for extended target tracking applied to X-band marine radar data,”IEEE Journal of Oceanic Engineering, vol. 41, no. 4, pp. 1007–1019, 2016

    work page 2016

  12. [12]

    Linear-time joint proba- bilistic data association for multiple extended object tracking,

    S. Yang, K. Thormann, and M. Baum, “Linear-time joint proba- bilistic data association for multiple extended object tracking,” in 10th Sensor Array and Multichannel Signal Processing Workshop. IEEE, 2018, pp. 6–10

    work page 2018

  13. [13]

    Scalable data association for extended object tracking,

    F. Meyer and M. Z. Win, “Scalable data association for extended object tracking,”IEEE Transactions on Signal and Information Processing over Networks, vol. 6, pp. 491–507, 2020

    work page 2020

  14. [14]

    Scalable detection and tracking of geo- metric extended objects,

    F. Meyer and J. Williams, “Scalable detection and tracking of geo- metric extended objects,”IEEE Transactions on Signal Processing, vol. 69, no. 6283–6298, 2021

    work page 2021

  15. [15]

    Multisensor multiple extended objects tracking based on the message passing,

    Y . Li, T. Shen, and L. Gao, “Multisensor multiple extended objects tracking based on the message passing,”IEEE Sensors Journal, vol. 24, no. 10, pp. 16 510–16 528, 2024

    work page 2024

  16. [16]

    Gaussian belief propagation based multi-view multi-extended target tracking with occlusion,

    Y . Guo, H. Zhang, B. Lin, H. Su, and Y . Chen, “Gaussian belief propagation based multi-view multi-extended target tracking with occlusion,”IEEE Sensors Journal, 2025

    work page 2025

  17. [17]

    Closed-form message passing algorithms for tracking extended targets,

    W. Ma, Z. Jing, P. Dong, and H. Leung, “Closed-form message passing algorithms for tracking extended targets,”IEEE Transac- tions on Aerospace and Electronic Systems, 2026

    work page 2026

  18. [18]

    Max sum based data associations for tracking point and extended targets,

    ——, “Max sum based data associations for tracking point and extended targets,”IEEE Transactions on Aerospace and Electronic Systems, 2024

    work page 2024

  19. [19]

    Extended target tracking using a Gaussian-mixture PHD filter,

    K. Granstr ¨om, C. Lundquist, and O. Orguner, “Extended target tracking using a Gaussian-mixture PHD filter,”IEEE Transactions on Aerospace and Electronic Systems, vol. 48, no. 4, pp. 3268– 3286, 2012

    work page 2012

  20. [20]

    An extended target CPHD filter and a gamma Gaussian inverse Wishart implementa- tion,

    C. Lundquist, K. Granstr ¨om, and U. Orguner, “An extended target CPHD filter and a gamma Gaussian inverse Wishart implementa- tion,”IEEE Journal of Selected Topics in Signal Processing, vol. 7, no. 3, pp. 472–483, 2013

    work page 2013

  21. [21]

    Multiple extended target tracking with labeled random finite sets,

    M. Beard, S. Reuter, K. Granstr ¨om, B.-T. V o, B.-N. V o, and A. Scheel, “Multiple extended target tracking with labeled random finite sets,”IEEE Transactions on Signal Processing, vol. 64, no. 7, pp. 1638–1653, 2016

    work page 2016

  22. [22]

    Poisson multi- Bernoulli mixture conjugate prior for multiple extended target fil- tering,

    K. Granstr ¨om, M. Fatemi, and L. Svensson, “Poisson multi- Bernoulli mixture conjugate prior for multiple extended target fil- tering,”IEEE Transactions on Aerospace and Electronic Systems, vol. 56, no. 1, pp. 208–225, 2020

    work page 2020

  23. [23]

    Poisson multi-Bernoulli approxi- mations for multiple extended object filtering,

    Y . Xia, K. Granstr ¨om, L. Svensson, M. Fatemi, ´A. F. Garc ´ıa- Fern´andez, and J. L. Williams, “Poisson multi-Bernoulli approxi- mations for multiple extended object filtering,”IEEE Transactions on Aerospace and Electronic Systems, vol. 58, no. 2, pp. 890–906, 2022

    work page 2022

  24. [24]

    Trajectory PMB filters for extended object tracking using belief propagation,

    Y . Xia, ´A. F. Garc ´ıa-Fern´andez, F. Meyer, J. L. Williams, K. Granstr ¨om, and L. Svensson, “Trajectory PMB filters for extended object tracking using belief propagation,”IEEE Trans- actions on Aerospace and Electronic Systems, 2023

    work page 2023

  25. [25]

    A density-based algorithm for discovering clusters in large spatial databases with noise

    M. Ester, H.-P. Kriegel, J. Sander, X. Xuet al., “A density-based algorithm for discovering clusters in large spatial databases with noise.” inKdd, vol. 96, no. 34, 1996, pp. 226–231

    work page 1996

  26. [26]

    On implementing 2D rectangular assignment algo- rithms,

    D. F. Crouse, “On implementing 2D rectangular assignment algo- rithms,”IEEE Transactions on Aerospace and Electronic Systems, vol. 52, no. 4, pp. 1679–1696, 2016

    work page 2016

  27. [27]

    Poisson multi-Bernoulli mapping using Gibbs sam- pling,

    M. Fatemi, K. Granstr ¨om, L. Svensson, F. J. Ruiz, and L. Ham- marstrand, “Poisson multi-Bernoulli mapping using Gibbs sam- pling,”IEEE Transactions on Signal Processing, vol. 65, no. 11, pp. 2814–2827, 2017

    work page 2017

  28. [28]

    Likelihood-based data association for extended object tracking using sampling methods,

    K. Granstr ¨om, L. Svensson, S. Reuter, Y . Xia, and M. Fatemi, “Likelihood-based data association for extended object tracking using sampling methods,”IEEE Transactions on Intelligent V ehi- cles, vol. 3, no. 1, pp. 30–45, 2018

    work page 2018

  29. [29]

    Variational tracking and redetection for closely-spaced objects in heavy clutter,

    R. Gan, Q. Li, and S. J. Godsill, “Variational tracking and redetection for closely-spaced objects in heavy clutter,”IEEE Transactions on Aerospace and Electronic Systems, vol. 60, no. 4, pp. 5286–5311, 2024

    work page 2024

  30. [30]

    Variational Bayesian inference for multiple extended targets or unresolved group targets tracking,

    Y . Cheng, Y . Cao, T.-S. Yeo, Y . Zhang, and J. Fu, “Variational Bayesian inference for multiple extended targets or unresolved group targets tracking,”IET Radar , Sonar & Navigation, vol. 19, no. 1, p. e70098, 2025

    work page 2025

  31. [31]

    Three- dimensional multiple extended targets tracking under occlusion using variational Gaussian processes,

    D. Yang, Y . Guo, X. Li, Y . Chen, and H. Shentu, “Three- dimensional multiple extended targets tracking under occlusion using variational Gaussian processes,”IEEE Transactions on Aerospace and Electronic Systems, vol. 61, no. 5, pp. 12 951– 12 969, 2025

    work page 2025

  32. [32]

    Pivot: Poisson measurements-based variational multi-object detection and tracking,

    R. Gan, Q. Li, J. R. Hopgood, M. E. Davies, and S. Godsill, “Pivot: Poisson measurements-based variational multi-object detection and tracking,” in28th International Conference on Information Fusion (FUSION). IEEE, 2025, pp. 1–8

    work page 2025

  33. [33]

    An adaptive and scal- able multi-object tracker based on the non-homogeneous Poisson process,

    Q. Li, R. Gan, J. Liang, and S. J. Godsill, “An adaptive and scal- able multi-object tracker based on the non-homogeneous Poisson process,”IEEE Transactions on Signal Processing, vol. 71, pp. 105–120, 2023

    work page 2023

  34. [34]

    A scalable Rao-Blackwellised sequential MCMC sampler for joint detection and tracking in clutter,

    Q. Li, R. Gan, and S. Godsill, “A scalable Rao-Blackwellised sequential MCMC sampler for joint detection and tracking in clutter,” in26th International Conference on Information Fusion (FUSION). IEEE, 2023, pp. 1–8

    work page 2023

  35. [35]

    A Poisson multi-Bernoulli mixture filter for coexisting point and extended targets,

    ´A. F. Garc´ıa-Fern´andez, J. L. Williams, L. Svensson, and Y . Xia, “A Poisson multi-Bernoulli mixture filter for coexisting point and extended targets,”IEEE Transactions on Signal Processing, vol. 69, pp. 2600–2610, 2021

    work page 2021

  36. [36]

    Poisson multi- Bernoulli mixture filter with general target-generated measure- ments and arbitrary clutter,

    ´A. F. Garc´ıa-Fern´andez, Y . Xia, and L. Svensson, “Poisson multi- Bernoulli mixture filter with general target-generated measure- ments and arbitrary clutter,”IEEE Transactions on Signal Pro- cessing, vol. 71, pp. 1895–1906, 2023

    work page 1906

  37. [37]

    Poisson multi-Bernoulli mixture filter: direct deriva- tion and implementation,

    ´A. F. Garc ´ıa-Fern´andez, J. L. Williams, K. Granstr ¨om, and L. Svensson, “Poisson multi-Bernoulli mixture filter: direct deriva- tion and implementation,”IEEE Transactions on Aerospace and Electronic Systems, vol. 54, no. 4, pp. 1883–1901, 2018

    work page 1901

  38. [38]

    Poisson multi-Bernoulli mixtures for sets of trajec- tories,

    K. Granstr ¨om, L. Svensson, Y . Xia, J. Williams, and ´A. F. Garc´ıa- Fern´andez, “Poisson multi-Bernoulli mixtures for sets of trajec- tories,”IEEE Transactions on Aerospace and Electronic Systems, vol. 61, no. 2, pp. 5178–5194, 2024

    work page 2024

  39. [39]

    An efficient implementation of the extended object trajectory PMB filter using blocked Gibbs sampling,

    Y . Xia, ´A. F. Garc ´ıa-Fern´andez, and L. Svensson, “An efficient implementation of the extended object trajectory PMB filter using blocked Gibbs sampling,” in26th International Conference on Information Fusion (FUSION). IEEE, 2023, pp. 1–8

    work page 2023

  40. [40]

    Gaussian implementation of the multi-Bernoulli mixture filter,

    ´A. F. Garc ´ıa-Fern´andez, Y . Xia, K. Granstr ¨om, L. Svensson, and J. L. Williams, “Gaussian implementation of the multi-Bernoulli mixture filter,” in22th International Conference on Information Fusion. IEEE, 2019, pp. 1–8

    work page 2019

  41. [41]

    LiDAR point cloud-based multiple vehicle tracking with probabilistic measurement-region association,

    G. Ding, J. Liu, Y . Xia, T. Huang, B. Zhu, and J. Sun, “LiDAR point cloud-based multiple vehicle tracking with probabilistic measurement-region association,” in27th International Confer- ence on Information Fusion (FUSION). IEEE, 2024, pp. 1–8

    work page 2024

  42. [42]

    A multiple extended object tracker with the Gaussian process model utilizing negative information,

    M. Baerveldt, M. E. L ´opez, and E. F. Brekke, “A multiple extended object tracker with the Gaussian process model utilizing negative information,”Journal of Advances in Information Fusion, vol. 19, no. 1, pp. 88–108, 2024

    work page 2024

  43. [43]

    Which framework is suitable for online 3D multi-object tracking for autonomous driving with automotive 4D imaging radar?

    J. Liu, G. Ding, Y . Xia, J. Sun, T. Huang, L. Xie, and B. Zhu, “Which framework is suitable for online 3D multi-object tracking for autonomous driving with automotive 4D imaging radar?” in IEEE Intelligent V ehicles Symposium (IV). IEEE, 2024, pp. 1258– 1265

    work page 2024

  44. [44]

    3-D multiple extended object tracking by fusing roadside radar and camera sensors,

    J. Deng, Z. Hu, Z. Lu, and X. Wen, “3-D multiple extended object tracking by fusing roadside radar and camera sensors,”IEEE Sensors Journal, vol. 25, no. 1, pp. 1885–1899, 2024

    work page 2024

  45. [45]

    Bayesian simulta- neous localization and multi-lane tracking using onboard sensors and a SD map,

    Y . Xia, E. Stenborg, J. Fu, and G. Hendeby, “Bayesian simulta- neous localization and multi-lane tracking using onboard sensors and a SD map,” in27th International Conference on Information Fusion (FUSION). IEEE, 2024, pp. 1–8

    work page 2024

  46. [46]

    5G SLAM using the clustering and assignment approach with diffuse multipath,

    Y . Ge, F. Wen, H. Kim, M. Zhu, F. Jiang, S. Kim, L. Svensson, and H. Wymeersch, “5G SLAM using the clustering and assignment approach with diffuse multipath,”Sensors, vol. 20, no. 16, p. 4656, 2020

    work page 2020

  47. [47]

    Combining oc- cupancy grid mapping and extended object tracking with the Poisson multi-Bernoulli mixture filter,

    M. Baerveldt, L. Svensson, and E. F. Brekke, “Combining oc- cupancy grid mapping and extended object tracking with the Poisson multi-Bernoulli mixture filter,”IEEE Journal of Oceanic Engineering, 2026

    work page 2026

  48. [48]

    B 2F-map: Crowd-sourced mapping with Bayesian B-spline fusion,

    Y . Xie, Y . Xia, E. Stenborg, J. Fu, A. Beauvisage, G. E. Garcia, T. Wu, and G. Hendeby, “B 2F-map: Crowd-sourced mapping with Bayesian B-spline fusion,” inProceedings of the IEEE International Conference on Robotics and Automation (ICRA), 2026, accepted. arXiv:2603.01673

    work page arXiv 2026

  49. [49]

    R. P. Mahler,Advances in Statistical Multisource-Multitarget Information Fusion. Artech House Norwood, MA, 2014

    work page 2014

  50. [50]

    Marginal multi-Bernoulli filters: RFS derivation of MHT, JIPDA, and association-based member,

    J. L. Williams, “Marginal multi-Bernoulli filters: RFS derivation of MHT, JIPDA, and association-based member,”IEEE Transactions on Aerospace and Electronic Systems, vol. 51, no. 3, pp. 1664– 1687, 2015

    work page 2015

  51. [51]

    Trajectory Poisson multi-Bernoulli filters,

    ´A. F. Garc´ıa-Fern´andez, L. Svensson, J. L. Williams, Y . Xia, and K. Granstr ¨om, “Trajectory Poisson multi-Bernoulli filters,”IEEE Transactions on Signal Processing, vol. 68, pp. 4933–4945, 2020

    work page 2020

  52. [52]

    Partially collapsed Gibbs samplers: Theory and methods,

    D. A. Van Dyk and T. Park, “Partially collapsed Gibbs samplers: Theory and methods,”Journal of the American Statistical Associ- ation, vol. 103, no. 482, pp. 790–796, 2008

    work page 2008

  53. [53]

    Koller and N

    D. Koller and N. Friedman,Probabilistic graphical models: prin- ciples and techniques. MIT press, 2009

    work page 2009

  54. [54]

    Sonar tracking of multiple targets using joint probabilistic data association,

    T. Fortmann, Y . Bar-Shalom, and M. Scheffe, “Sonar tracking of multiple targets using joint probabilistic data association,”IEEE journal of Oceanic Engineering, vol. 8, no. 3, pp. 173–184, 1983

    work page 1983

  55. [55]

    Extended target Poisson multi-Bernoulli mixture trackers based on sets of trajectories,

    Y . Xia, K. Granstr ¨om, L. Svensson, ´A. F. Garc ´ıa-Fern´andez, and J. L. Williams, “Extended target Poisson multi-Bernoulli mixture trackers based on sets of trajectories,” in22th International Con- ference on Information Fusion. IEEE, 2019, pp. 1–8

    work page 2019

  56. [56]

    Tracking of extended objects and group targets using random matrices,

    M. Feldmann, D. Franken, and W. Koch, “Tracking of extended objects and group targets using random matrices,”IEEE Transac- tions on Signal Processing, vol. 59, no. 4, pp. 1409–1420, 2011

    work page 2011

  57. [57]

    Estimation and maintenance of measurement rates for multiple extended target tracking,

    K. Granstr ¨om and U. Orguner, “Estimation and maintenance of measurement rates for multiple extended target tracking,” in Proceedings of International Conference on Information Fusion. IEEE, 2012, pp. 2170–2176

    work page 2012

  58. [58]

    Gelman, J

    A. Gelman, J. B. Carlin, H. S. Stern, and D. B. Rubin,Bayesian data analysis. Chapman and Hall/CRC, 1995

    work page 1995

  59. [59]

    On the reduction of Gaussian inverse Wishart mixtures,

    K. Granstr ¨om and U. Orguner, “On the reduction of Gaussian inverse Wishart mixtures,” inInternational Conference on Infor- mation Fusion. IEEE, 2012, pp. 2162–2169

    work page 2012

  60. [60]

    Scalable detection and tracking of extended objects,

    F. Meyer and J. L. Williams, “Scalable detection and tracking of extended objects,” inIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2020, pp. 8916– 8920

    work page 2020

  61. [61]

    Generalized optimal sub-pattern assignment metric,

    A. S. Rahmathullah, ´A. F. Garc ´ıa-Fern´andez, and L. Svensson, “Generalized optimal sub-pattern assignment metric,” inProceed- ings of International Conference on Information Fusion. IEEE, 2017, pp. 1–8

    work page 2017

  62. [62]

    Metrics for performance evaluation of elliptic extended object tracking methods,

    S. Yang, M. Baum, and K. Granstr ¨om, “Metrics for performance evaluation of elliptic extended object tracking methods,” inInter- national Conference on Multisensor Fusion and Integration for Intelligent Systems. IEEE, 2016, pp. 523–528. : 17

    work page 2016