Pith. sign in

REVIEW 3 major objections 4 minor 69 references

Robust Online Multi-target Visual Tracking using a HISP Filter with Discriminative Deep Appearance Learning

T0 review · 3 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A HISP filter with a deep appearance likelihood lifts MOTA from 35.9 to 37.4 on MOT16 and to 45.4 on MOT17 while keeping track identities online.

desk verdict A plausible HISP-plus-deep-appearance integration with modest but consistent MOT gains, undercut by an underspecified likelihood and no code. read the letter →

arxiv 1908.03945 v6 pith:PJ2KWWQK submitted 2019-08-11 cs.CV

Nathanael L. Baisa This is my paper · ORCID
classification cs.CV
keywords HISPfilterstochasticpopulationsmulti-targettrackingdeepappearancelearningtracking-by-detectionpersonre-identificationMOTbenchmarkaugmentedlikelihood
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 tries to establish that the HISP filter, a stochastic-population multi-target filter that carries track identities at linear computational cost, can serve as an online visual tracker when its measurement likelihood is augmented with deep CNN appearance features. The author trains a Siamese ResNet50 verification-identification network (VerIdNet) on 6,654 person identities and combines the cosine distance of its features with a Gaussian motion likelihood inside the filter's hypothesis weights. The central reported evidence is that the deep-appearance version (HISP-DAL) raises MOTA from 35.9 to 37.4 and IDF1 from 28.9 to 30.5 on MOT16 relative to the motion-only HISP-T, and reaches 45.4 MOTA on MOT17 with public detections. If correct, this is the first HISP-filter tracker with deep representation learning, and it shows that appearance cues can enter stochastic-population filters through the likelihood rather than through a separate association stage.

What carries the argument

The load-bearing object is the HISP filter itself, a Bayesian multi-target filter that represents each previously detected target by a measurement path, or hypothesis, and propagates each hypothesis' probability density and survival weight with complexity linear in the number of hypotheses and measurements. Into this filter the paper inserts an augmented likelihood of the form $g_t(z|x)=g_t(z_d|x)g_t(z_w|x)$: the detection term is the Gaussian innovation of Eq. (12), and the appearance term is the sigmoid of the cosine distance between CNN features, Eq. (21). The VerIdNet supplies those features: a Siamese ResNet50 with one verification and two identification losses, trained on 6,654 identities. The filter's update equations then convert these likelihoods into association weights and posterior hypothesis weights, so the appearance term is used both for data association and for track updates.

What would settle it

Rerun HISP-DAL on the MOT16 test set with the appearance factor $g_t(z_w\mid x)$ in Eq. (21) replaced by the constant 1: if MOTA remains 37.4, the deep appearance term is not carrying the reported improvement.

Watch

Extended reading notes

Core claim

The paper's central assertion is that the HISP filter's per-track hypotheses can ingest an augmented likelihood $g_t(z|x)=g_t(z_d|x)g_t(z_w|x)$, where the first factor scores box-level agreement through the Kalman innovation and the second factor scores appearance agreement through a logistic function of the cosine distance between deep CNN embeddings. With this likelihood, HISP-DAL reports 37.4 MOTA and 2,101 identity switches on MOT16 versus 35.9 MOTA and 2,592 switches for the motion-only HISP-T, and reaches 45.4 MOTA on MOT17. The paper also claims that the weight propagated with each confirmed hypothesis can relabel spawned objects that would otherwise share one identity after track extraction. The appearance embeddings come from VerIdNet, a Siamese ResNet50 trained with two identification losses and one verification loss on a mixture of re-identification and tracking data sets, which the paper reports at about 98% pair-verification accuracy.

Load-bearing premise

The augmented likelihood can be evaluated only if every track has an appearance feature vector available at its predicted state, but the paper never defines how such a track appearance template is initialized, stored, or updated.

Editorial extensions

If this is right

  • An online tracker can preserve track identities without exponential hypothesis growth; the reported complexity is $O(|\mathcal{I}_{t|t-1}||Z_t|)$ per recursion.
  • Appearance learning can be added to stochastic-population filters through the measurement likelihood, so the same augmented likelihood should extend to DISP and related filters, not only HISP.
  • The paper's reported MOTA on MOT17 is obtained with public detections from DPM, FRCNN, and SDP detectors, so the gain is not tied to a custom detector.
  • The duplicate-label fix is applied after track extraction and does not alter the filtering recursion, which means any HISP implementation can adopt it without changing the filter itself.

Reading between the lines

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

  • The paper leaves a track's appearance state undefined: nothing in the Kalman state of Section V stores or updates an appearance template, so a working implementation must add an undocumented appearance store. Making that store explicit would also make Eq. (13) testable.
  • The same cosine-sigmoid likelihood could be ported to other Bayesian multi-target filters, such as labeled random-finite-set filters; if the reported gain is real, those filters should see a similar improvement.
  • The VerIdNet's 98% pair-verification accuracy contrasts with the tracker's 30.5 IDF1 on MOT16, suggesting that the appearance signal is diluted by the motion term or by association ambiguity; a testable extension is to reweight $g_t(z_d|x)$ and $g_t(z_w|x)$.
  • Since HISP-DAL is compared only with the author's own motion-only HISP-T, an independent appearance-based tracker under identical detections would be needed to isolate the contribution of the deep features.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper proposes an online multi-target visual tracker based on the Hypothesized and Independent Stochastic Population (HISP) filter, combined with a deep CNN appearance model (VerIdNet). The authors augment the HISP measurement likelihood with an appearance term, g_t(z|x)=g_t(z_d|x)g_t(z_w|x), and evaluate the resulting tracker (HISP-DAL) on MOT16 and MOT17. They report MOTA of 37.4 on MOT16 (versus 35.9 for the motion-only HISP-T) and 45.4 on MOT17, and claim the deep appearance information is responsible for the improvement. The paper also introduces a heuristic to re-label spawned tracks that share an identity after track extraction.

Significance. If the central claim is correct, this is the first HISP filter with deep representation learning and a working online identity-preserving tracker, which would be a useful contribution to visual multi-target tracking: the HISP filter offers linear complexity while maintaining track identities, and the addition of discriminative appearance features is a natural and potentially valuable extension. The benchmark results are self-consistent, and the use of external MOT16/MOT17 evaluation reduces circularity concerns. However, the current manuscript does not fully specify the mechanism by which the deep appearance likelihood is evaluated, and one key equation is mathematically problematic, so the central claim is not yet verifiable from the text.

major comments (3)
  1. [§III-B and §IV-A] The augmented likelihood g_t(z_w|x) in Eq. (21) cannot be evaluated as written because the paper never defines a track appearance state. Section V defines the Kalman state as x_t=[p_cx, p_cy, v_x, v_y, w, h]^T and the measurement as [p_cx, p_cy, w, h]^T, with no appearance descriptor. Eq. (21) requires z_i_w, an appearance feature extracted from the detected object region determined by the predicted measurement Hx, but no template store, no initialization rule, and no update rule for an appearance template are given. Please specify how the appearance feature of a predicted track is obtained: is a crop taken from the current frame at the predicted box, or is a stored template used? How is the template initialized at track birth and updated after association, and how are miss-detections handled? Without this, the reported MOTA gain from deep appearance cannot be traced to the equations provided.
  2. [§III-B, Eq. (13)] Equation (13) is not a valid identity. The left-hand side is an integral over x, but the right-hand side retains Hx and g_t(z_w|x) after the integral. For a Gaussian p(x), the detection-term integral should be evaluated at the predicted mean Hm, and the appearance term, if it depends on x, must remain inside the integral or be approximated explicitly. As written, the equation cannot be used to compute the association weight w^{κ,z}_t. Please correct the derivation and state the approximations used to factor the appearance term out of the integral.
  3. [§VI, Ablation Study] The ablation compares HISP-DAL with HISP-T, which is reported in the author's prior work [22], rather than a re-run of the same implementation without the appearance term. The paper states that HISP-T is 'simply the HISP-DAL tracker without deep appearance information,' but no evidence is provided that all other components (Kalman parameters, pruning/merging thresholds, track extraction window, detection preprocessing) are identical. The 1.5 MOTA difference could stem from implementation differences or parameter changes. Please provide a controlled ablation: run the same code with the appearance likelihood disabled, or report the exact configuration differences between HISP-DAL and HISP-T.
minor comments (4)
  1. [Abstract and §VI] The abstract claims the tracker 'significantly outperforms several state-of-the-art trackers in terms of tracking accuracy,' but on MOT16, AMIR and KCF16 achieve higher MOTA (47.2 and 48.8 versus 37.4, Table I). The claim is acceptable if 'several' is interpreted as 'many, but not all,' but it would be clearer to qualify the statement or highlight the MOT17 results where the tracker does rank first in MOTA.
  2. [§III-B, Eq. (12)] The detection likelihood in Eq. (12) is written as an unnormalized exponential. Since the text states that the augmented likelihood is 'dimensionless and takes values in the interval [0,1]', the normalization is not just cosmetic; please define whether g_t(z_d|x) is a likelihood ratio, a density, or an unnormalized score, and clarify how it is normalized.
  3. [§IV-A, Eqs. (21) and (22)] Eq. (22) defines C_w as a cosine similarity (dot product divided by norms), but the text calls it a cosine distance. A similarity of +1 corresponds to identical features, while a distance would be 0 in that case. Please make the terminology consistent and state the range of C_w, since Eq. (21) is a sigmoid of 2C_w and the interpretation depends on the sign convention.
  4. [§III-D] The proposed re-labeling of spawned tracks based on the maximum-weight assumption is described as rarely susceptible to identity switches, but no quantitative evidence supports this claim. A brief analysis of how often the assumption fails (e.g., on the reported sequences) would help the reader assess the robustness of this post-processing step.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the central tracking-accuracy gain is evaluated on external MOT16/MOT17 test sets with fixed parameters and a separately trained appearance network.

full rationale

The paper's central claim, that incorporating deep CNN appearance features through the augmented likelihood g_t(z|x)=g_t(z_d|x)g_t(z_w|x) improves tracking, is supported by an empirical comparison on the external MOT16/MOT17 test sets using public detections. The reported MOTA values are not fitted to the result: the filter parameters (sigma_v=5, sigma_r=6, p_d=0.9, clutter rate 10, n_alpha=0.1) are stated as fixed implementation values, and the VerIdNet appearance model is trained on person re-identification data rather than on the MOT test sequences. The comparisons against HISP-T [22] and GM-PHD-DAL [3] cite the author's own prior publications, but those are used only as published baselines; the improvement claim is not derived from those citations by construction. The HISP filter itself is taken from external work [14]-[16], not from any uniqueness theorem of the present author. The only notable defect is that Eq. (13) leaves an appearance-feature term depending on x outside the integral, and no track appearance template or update rule is defined in Section V, so the augmented likelihood is under-specified as written. That is an implementation/correctness gap, not a circular reduction: no equation is defined in terms of the output it is claimed to predict, and no fitted quantity is relabeled as a prediction. Therefore no circularity is found, and the score is 0.

Assumptions & free parameters 7 free parameters · 4 assumptions · 0 invented entities

No new physical or hypothetical entities are introduced; VerIdNet is a neural network architecture, not a postulated entity. Everything in the ledger is either a hand-set parameter, an inherited filter assumption, or an unstated model for track appearance.

free parameters (7)
  • Probability of detection p_d = 0.90
    Set as a constant in Section V; it controls the measurement update in Eq. (10) and how many missed detections a track can survive.
  • Target survival probability p_pi = 0.99 (death kernel q_w = 1e-2)
    Set in Section V; the paper states the HISP filter is sensitive to this value and that it must exceed p_d to handle miss-detections.
  • False alarm (clutter) rate = mean 10 per frame; v_t(z) ~ 4.8e-6
    Poisson clutter model in Section V; it enters the association weights in Eqs. (16)-(18).
  • Birth intensity n_alpha = 0.1 new objects per frame
    Set in Section V; controls how readily new tracks are created.
  • Process and observation noise = sigma_v = 5 px/s^2, sigma_r = 6 px
    Kalman noise parameters in Eqs. (23)-(24); chosen by hand.
  • Pruning, merging, and window thresholds = tau_p = 1e-3, tau_m = 4 px, T = 5, max hypotheses = 1e7
    Track management settings in Section V that determine which hypotheses survive to extraction.
  • Cosine verification threshold = 0.75
    Threshold used in VerIdNet evaluation in Section IV-B; it also sets the operating point of the appearance likelihood.
assumptions (4)
  • domain assumption The HISP filter recursion and its linear complexity are taken as given from the stochastic populations literature.
    The paper applies refs [14]-[16] without deriving the filter; the tracker's correctness and complexity claims inherit from that theory.
  • domain assumption Data association is moderately ambiguous, so products of association weights for different hypotheses vanish (sparsity assumption).
    Stated in Section III-B after Eq. (14). Dense pedestrian scenes such as MOT17-07 may violate this.
  • ad hoc to paper The appearance likelihood can be evaluated as a function of a predicted kinematic state, i.e., a track appearance template exists and its cosine distance with the current detection can be computed.
    Eqs. (11)-(13) and Section IV-A imply that z_i_w exists for the predicted measurement Hx, but Section V defines no appearance state or template update.
  • domain assumption MOT16/MOT17 public detections and ground truth are reliable and the benchmark evaluation is unbiased.
    All experimental claims rest on the standard MOT benchmark protocol with public detections.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Robust Online Multi-target Visual Tracking using a HISP Filter with Discriminative Deep Appearance Learning." pith.science (2026). https://pith.science/paper/PJ2KWWQK

@misc{pith2026190803945,
  author       = {Pith},
  title        = {Pith review of: Robust Online Multi-target Visual Tracking using a HISP Filter with Discriminative Deep Appearance Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PJ2KWWQK}},
  note         = {Machine review of arXiv:1908.03945}
}
read the original abstract

We propose a novel online multi-target visual tracker based on the recently developed Hypothesized and Independent Stochastic Population (HISP) filter. The HISP filter combines advantages of traditional tracking approaches like MHT and point-process-based approaches like PHD filter, and it has linear complexity while maintaining track identities. We apply this filter for tracking multiple targets in video sequences acquired under varying environmental conditions and targets density using a tracking-by-detection approach. We also adopt deep CNN appearance representation by training a verification-identification network (VerIdNet) on large-scale person re-identification data sets. We construct an augmented likelihood in a principled manner using this deep CNN appearance features and spatio-temporal information. Furthermore, we solve the problem of two or more targets having identical label considering the weight propagated with each confirmed hypothesis. Extensive experiments on MOT16 and MOT17 benchmark data sets show that our tracker significantly outperforms several state-of-the-art trackers in terms of tracking accuracy.

Figures

Figures reproduced from arXiv: 1908.03945 by the authors.

Figure 2
Figure 2. The transition functions’ relation between times t-1 and t for the subsets of X. 1A(x) = ( 1, if x ∈ A 0, otherwise (3) Using the previously mentioned concept and notations, the HISP filter can be described via a set of hypotheses. For example, after the measurement update step at time t (see section III-B), it can be described by set of triples (multi￾target configuration) of the form Pt = {p i t , wi t , ni t}i∈It… view at source ↗
Figure 3
Figure 3. Illustration of VerIdNet for discriminative deep appearance learning using two ResNet50 models. [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Sample results on several sequences of MOT17 data sets using SDP detector; bounding boxes represent the tracking results with their color-coded identities. From left to right: MOT17-01-SDP and MOT17-03-SDP (top row), MOT17-06-SDP and MOT17-08-SDP (middle row), and MOT17-12-SDP and MOT17-14-SDP (bottom row). The videos of tracking results are available on the MOT Challenge website https://motchallenge.net/. into a si… view at source ↗
Figures from the paper (1 more)
Figure 5
Figure 5. Figure 5: Sample results on the sequence MOT17-07-SDP (using SDP detector); bounding boxes represent the tracking results with their color-coded identities, for frames 376, 386 and 395 from top to bottom. The video of tracking results are available on the MOT Challenge website h…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

69 extracted references · 61 canonical work pages

  1. [22]

    Online multi-target visual tracking using a HISP filter,

    N. L. Baisa, “Online multi-target visual tracking using a HISP filter,” in Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP ,, INSTICC. SciTePress, 2018, pp. 429–438

    work page 2018

  2. [1]

    Online multi-target tracking with strong and weak detections,

    R. Sanchez-Matilla, F. Poiesi, and A. Cavallaro, “Online multi-target tracking with strong and weak detections,” in Computer Vision - ECCV 2016 Workshops - Amsterdam, The Netherlands, October 8-10 and 15- 16, 2016, Proceedings, Part II , 2016, pp. 84–99. 13 Fig. 5: Sample results on the sequence MOT17-07-SDP (using SDP detector); bounding boxes represent ...

    work page 2016

  3. [2]

    Online multiple object tracking with the hierarchically adopted GM-PHD filter using motion and appearance,

    Y . Song and M. Jeon, “Online multiple object tracking with the hierarchically adopted GM-PHD filter using motion and appearance,” in IEEE/IEIE The International Conference on Consumer Electronics (ICCE) Asia, 2016

    work page 2016

  4. [3]

    Online multi-object visual tracking using a GM-PHD filter with deep appearance learning,

    N. L. Baisa, “Online multi-object visual tracking using a GM-PHD filter with deep appearance learning,” in 2019 22nd International Conference on Information Fusion (FUSION) , July 2019

    work page 2019

  5. [4]

    Learning by track- ing: Siamese CNN for robust target association,

    L. Leal-Taix ´e, C. Canton-Ferrer, and K. Schindler, “Learning by track- ing: Siamese CNN for robust target association,” IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPR). DeepVi- sion: Deep Learning for Computer Vision. , 2016

    work page 2016

  6. [5]

    Continuous energy minimization for multitarget tracking,

    A. Milan, S. Roth, and K. Schindler, “Continuous energy minimization for multitarget tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, no. 1, pp. 58–72, Jan 2014

    work page 2014

  7. [6]

    Globally-optimal greedy algorithms for tracking a variable number of objects,

    H. Pirsiavash, D. Ramanan, and C. C. Fowlkes, “Globally-optimal greedy algorithms for tracking a variable number of objects,” in CVPR 2011, June 2011, pp. 1201–1208

    work page 2011

  8. [7]

    B.-N. V o, M. Mallick, Y . Bar-Shalom, S. Coraluppi, R. O. III, R. Mahler, and B.-T. V o,”Multitarget Tracking”. Wiley Encyclopedia of Electrical and Electronics Engineering, Wiley, September 2015

    work page 2015

Show all 69 references
  1. [8]

    Multiple object tracking: A literature review,

    W. Luo, X. Zhao, and T. Kim, “Multiple object tracking: A literature review,”CoRR, vol. abs/1409.7618, 2014. [Online]. Available: http://arxiv.org/abs/1409.7618

    2014 arXiv

  2. [9]

    Bar-Shalom, P

    Y . Bar-Shalom, P. Willett, and X. Tian, Tracking and Data Fusion: A Handbook of Algorithms . YBS Publishing, 2011. [Online]. Available: https://books.google.co.uk/books?id=2aOiuAAACAAJ

    2011

  3. [10]

    Multiple hypothesis tracking revisited,

    C. Kim, F. Li, A. Ciptadi, and J. M. Rehg, “Multiple hypothesis tracking revisited,” in 2015 IEEE International Conference on Computer Vision (ICCV), Dec 2015, pp. 4696–4704

    2015

  4. [11]

    An extension of the munkres algorithm for the assignment problem to rectangular matrices,

    F. Bourgeois and J.-C. Lassalle, “An extension of the munkres algorithm for the assignment problem to rectangular matrices,” Commun. ACM , vol. 14, no. 12, pp. 802–804, Dec. 1971

    1971

  5. [12]

    R. P. Mahler, Advances in statistical multisource-multitarget information fusion. Norwood: Artech House, 2014. [Online]. Available: http: //cds.cern.ch/record/2158275

    2014

  6. [13]

    Multitarget bayes filtering via first-order multitarget moments,

    ——, “Multitarget bayes filtering via first-order multitarget moments,” IEEE Trans. on Aerospace and Electronic Systems , vol. 39, no. 4, pp. 1152–1178, 2003

    2003

  7. [14]

    Multitarget filtering with linearized complexity,

    J. Houssineau and D. E. Clark, “Multitarget filtering with linearized complexity,” IEEE Transactions on Signal Processing , vol. 66, no. 18, pp. 4957–4970, Sep. 2018

    2018

  8. [15]

    A new multi-target tracking algorithm for a large number of orbiting objects,

    E. Delande, J. Houssineau, J. Franco, C. Fr ¨uh, and D. Clark, “A new multi-target tracking algorithm for a large number of orbiting objects,” 27th AAS/AIAA Space Flight Mechanics Meeting , 2017

    2017

  9. [16]

    A new multi-target tracking algorithm for a large number of orbiting objects,

    E. Delande, J. Houssineau, J. Franco, C. Frueh, D. Clark, and M. Jah, “A new multi-target tracking algorithm for a large number of orbiting objects,” Advances in Space Research , vol. 64, no. 3, pp. 645 – 667,

  10. [17]

    ImageNet classification with deep convolutional neural networks,

    A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” in Advances in Neural In- formation Processing Systems 25 , F. Pereira, C. Burges, L. Bottou, and K. Weinberger, Eds., 2012, pp. 1097–1105

    2012

  11. [18]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” CoRR, vol. abs/1512.03385, 2015

    2015 arXiv

  12. [19]

    Faster R-CNN: towards 14 real-time object detection with region proposal networks,

    S. Ren, K. He, R. B. Girshick, and J. Sun, “Faster R-CNN: towards 14 real-time object detection with region proposal networks,” CoRR, vol. abs/1506.01497, 2015. [Online]. Available: http://arxiv.org/abs/1506. 01497

    2015 arXiv

  13. [20]

    A discriminatively learned CNN embedding for person re-identification,

    Z. Zheng, L. Zheng, and Y . Yang, “A discriminatively learned CNN embedding for person re-identification,” CoRR, vol. abs/1611.05666, 2016

    2016 arXiv

  14. [21]

    Tracking the untrackable: Learning to track multiple cues with long-term dependencies,

    A. Sadeghian, A. Alahi, and S. Savarese, “Tracking the untrackable: Learning to track multiple cues with long-term dependencies,” CoRR, vol. abs/1701.01909, 2017. [Online]. Available: http://arxiv.org/abs/ 1701.01909

    2017 arXiv

  15. [23]

    The Gaussian mixture probability hypothesis density filter,

    B.-N. V o and W.-K. Ma, “The Gaussian mixture probability hypothesis density filter,” Signal Processing, IEEE Transactions on, vol. 54, no. 11, pp. 4091–4104, Nov 2006

    2006

  16. [24]

    Long-term correlation tracking using multi-layer hybrid features in sparse and dense environments,

    N. L. Baisa, D. Bhowmik, and A. Wallace, “Long-term correlation tracking using multi-layer hybrid features in sparse and dense environments,” Journal of Visual Communication and Image Representation, vol. 55, pp. 464 – 476, 2018. [Online]. Available: http://www.sciencedirect.c...

    2018

  17. [25]

    Single to multiple target, multiple type visual tracking,

    N. L. Baisa, “Single to multiple target, multiple type visual tracking,” Ph.D. dissertation, Heriot-Watt University, 06 2018

    2018

  18. [26]

    Multiple target, multiple type filtering in the RFS framework,

    N. L. Baisa and A. Wallace, “Multiple target, multiple type filtering in the RFS framework,” Digital Signal Processing , vol. 89, pp. 49 – 59,

  19. [27]

    Development of a N-type GM-PHD filter for multiple target, multiple type visual tracking,

    ——, “Development of a N-type GM-PHD filter for multiple target, multiple type visual tracking,” Journal of Visual Communication and Image Representation, vol. 59, pp. 257 – 271, 2019. [Online]. Available: http://www.sciencedirect.com/science/article/pii/S1047320319300343

    2019

  20. [28]

    Available: http://www.sciencedirect.com/science/article/ pii/S1051200418303166

    [Online]. Available: http://www.sciencedirect.com/science/article/ pii/S1051200418303166

  21. [29]

    An efficient implementation of the generalized labeled multi-bernoulli filter,

    B. V o, B. V o, and H. G. Hoang, “An efficient implementation of the generalized labeled multi-bernoulli filter,” IEEE Transactions on Signal Processing, vol. 65, no. 8, pp. 1975–1987, April 2017

    1975

  22. [30]

    Labeled random finite sets and the bayes multi-target tracking filter,

    B. V o, B. V o, and D. Phung, “Labeled random finite sets and the bayes multi-target tracking filter,” IEEE Transactions on Signal Processing , vol. 62, no. 24, pp. 6554–6567, Dec 2014

    2014

  23. [31]

    Multi-object filtering with stochastic populations,

    E. Delande, J. Houssineau, and D. Clark, “Multi-object filtering with stochastic populations,” arXiv, vol. 1501.04671v2, 2016. [Online]. Available: https://arxiv.org/abs/1501.04671v2

    2016 arXiv

  24. [32]

    Online multi-object tracking via labeled random finite set with appearance learning,

    D. Y . Kim, “Online multi-object tracking via labeled random finite set with appearance learning,” in 2017 International Conference on Control, Automation and Information Sciences (ICCAIS), Oct 2017, pp. 181–186

    2017

  25. [33]

    Tracking interacting objects using intertwined flows,

    X. Wang, E. T ¨uretken, F. Fleuret, and P. Fua, “Tracking interacting objects using intertwined flows,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 38, no. 11, pp. 2312–2326, 2016

    2016

  26. [34]

    Multiple object tracking using k-shortest paths optimization,

    J. Berclaz, F. Fleuret, E. Turetken, and P. Fua, “Multiple object tracking using k-shortest paths optimization,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33, no. 9, pp. 1806–1819, 2011

    2011

  27. [35]

    Greedy batch-based minimum-cost flows for tracking multiple objects,

    X. Wang, B. Fan, S. Chang, Z. Wang, X. Liu, D. Tao, and T. S. Huang, “Greedy batch-based minimum-cost flows for tracking multiple objects,” IEEE Transactions on Image Processing, vol. 26, no. 10, pp. 4765–4776, 2017

    2017

  28. [36]

    Non-markovian globally consistent multi-object tracking,

    A. Maksai, X. Wang, F. Fleuret, and P. Fua, “Non-markovian globally consistent multi-object tracking,” in Proceedings of the IEEE Interna- tional Conference on Computer Vision (ICCV) , Oct 2017

    2017

  29. [37]

    GM-PHD filter based online multiple human tracking using deep discriminative corre- lation matching,

    Z. Fu, F. Angelini, S. M. Naqvi, and J. A. Chambers, “GM-PHD filter based online multiple human tracking using deep discriminative corre- lation matching,” in 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , April 2018, pp. 4299–4303

    2018

  30. [38]

    Multi-object tracking with neural gating using bilinear LSTM,

    C. Kim, F. Li, and J. M. Rehg, “Multi-object tracking with neural gating using bilinear LSTM,” in The European Conference on Computer Vision (ECCV), September 2018

    2018

  31. [39]

    Scalable person re-identification: A benchmark,

    L. Zheng, L. Shen, L. Tian, S. Wang, J. Wang, and Q. Tian, “Scalable person re-identification: A benchmark,” in 2015 IEEE International Conference on Computer Vision (ICCV) , Dec 2015, pp. 1116–1124

    2015

  32. [40]

    Deep learning face representation by joint identification-verification,

    Y . Sun, X. Wang, and X. Tang, “Deep learning face representation by joint identification-verification,” in NIPS, 2014

    2014

  33. [41]

    Stable multi-target tracking in real-time surveillance video,

    B. Benfold and I. Reid, “Stable multi-target tracking in real-time surveillance video,” in CVPR, June 2011, pp. 3457–3464

    2011

  34. [42]

    ImageNet Large Scale Visual Recognition Challenge,

    O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, A. C. Berg, and L. Fei-Fei, “ImageNet Large Scale Visual Recognition Challenge,” International Journal of Computer Vision (IJCV) , vol. 115, no. 3, pp. 211–252, 2015

    2015

  35. [43]

    Deepreid: Deep filter pairing neural network for person re-identification,

    W. Li, R. Zhao, T. Xiao, and X. Wang, “Deepreid: Deep filter pairing neural network for person re-identification,” in 2014 IEEE Conference on Computer Vision and Pattern Recognition , June 2014, pp. 152–159

    2014

  36. [44]

    Performance measures and a data set for multi-target, multi- camera tracking,

    E. Ristani, F. Solera, R. S. Zou, R. Cucchiara, and C. Tomasi, “Performance measures and a data set for multi-target, multi- camera tracking,” in Computer Vision - ECCV 2016 Workshops - Amsterdam, The Netherlands, October 8-10 and 15-16, 2016, Proceedings, Part II , 2016, pp. ...

    2016 doi

  37. [46]

    Region-based quality estimation network for large-scale person re-identification,

    G. Song, B. Leng, Y . Liu, C. Hetang, and S. Cai, “Region-based quality estimation network for large-scale person re-identification,” CoRR, vol. abs/1711.08766, 2017. [Online]. Available: http://arxiv.org/ abs/1711.08766

    2017 arXiv

  38. [47]

    MOT16: A benchmark for multi-object tracking,

    A. Milan, L. Leal-Taix ´e, I. Reid, S. Roth, and K. Schindler, “MOT16: A benchmark for multi-object tracking,” arXiv:1603.00831 [cs], Mar. 2016, arXiv: 1603.00831. [Online]. Available: http: //arxiv.org/abs/1603.00831

    2016 arXiv

  39. [48]

    Object detection with discriminatively trained part-based models,

    P. F. Felzenszwalb, R. B. Girshick, D. McAllester, and D. Ramanan, “Object detection with discriminatively trained part-based models,” IEEE Trans. Pattern Anal. Mach. Intell. , vol. 32, no. 9, pp. 1627–1645, Sep. 2010. [Online]. Available: http://dx.doi.org/10.1109/TPAMI.2009. 167

    2010 doi

  40. [49]

    A sequential monte carlo approximation of the HISP filter,

    J. Houssineau, D. E. Clark, and P. Del Moral, “A sequential monte carlo approximation of the HISP filter,” in Signal Processing Conference (EUSIPCO), 2015 23rd European . IEEE, 2015, pp. 1251–1255

    2015

  41. [50]

    Evaluating multiple object tracking performance: The CLEAR MOT metrics,

    K. Bernardin and R. Stiefelhagen, “Evaluating multiple object tracking performance: The CLEAR MOT metrics,” J. Image Video Process., pp. 1:1–1:10, Jan 2008

    2008

  42. [51]

    Learning to associate: Hybridboosted multi-target tracker for crowded scene,

    Y . Li, C. Huang, and R. Nevatia, “Learning to associate: Hybridboosted multi-target tracker for crowded scene,” in In CVPR, 2009

    2009

  43. [52]

    Exploit all the layers: Fast and accurate cnn object detector with scale dependent pooling and cascaded rejection classifiers,

    F. Yang, W. Choi, and Y . Lin, “Exploit all the layers: Fast and accurate cnn object detector with scale dependent pooling and cascaded rejection classifiers,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016, pp. 2129–2137

    2016

  44. [53]

    High-speed tracking-by- detection without using image information,

    E. Bochinski, V . Eiselein, and T. Sikora, “High-speed tracking-by- detection without using image information,” in 2017 14th IEEE Inter- national Conference on Advanced Video and Signal Based Surveillance (AVSS), Aug 2017, pp. 1–6

    2017

  45. [54]

    The way they move: Tracking multiple targets with similar appearance,

    C. Dicle, O. I. Camps, and M. Sznaier, “The way they move: Tracking multiple targets with similar appearance,” in 2013 IEEE International Conference on Computer Vision , Dec 2013, pp. 2304–2311

    2013

  46. [55]

    MatConvNet – convolutional neural networks for matlab,

    A. Vedaldi and K. Lenc, “MatConvNet – convolutional neural networks for matlab,” in Proceedings of the 25th annual ACM international conference on Multimedia , 2015

    2015

  47. [56]

    Sequential sensor fusion combining probability hypothesis density and kernelized correlation filters for multi-object tracking in video data,

    T. Kutschbach, E. Bochinski, V . Eiselein, and T. Sikora, “Sequential sensor fusion combining probability hypothesis density and kernelized correlation filters for multi-object tracking in video data,” in 2017 14th IEEE International Conference on Advanced Video and Signal Base...

    2017

  48. [57]

    Real-time multi- human tracking using a probability hypothesis density filter and multiple detectors,

    V . Eiselein, D. Arp, M. P ¨atzold, and T. Sikora, “Real-time multi- human tracking using a probability hypothesis density filter and multiple detectors,” in 2012 IEEE Ninth International Conference on Advanced Video and Signal-Based Surveillance , Sep. 2012, pp. 325–330

    2012

  49. [58]

    Joint probabilistic data association revisited,

    S. H. Rezatofighi, A. Milan, Z. Zhang, Q. Shi, A. Dick, and I. Reid, “Joint probabilistic data association revisited,” in 2015 IEEE Interna- tional Conference on Computer Vision (ICCV) , Dec 2015, pp. 3047– 3055

    2015

  50. [59]

    Eliminating exposure bias and metric mismatch in multiple object tracking,

    A. Maksai and P. Fua, “Eliminating exposure bias and metric mismatch in multiple object tracking,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , June 2019

    2019

  51. [60]

    Multiple object tracking via feature pyramid siamese networks,

    S. Lee and E. Kim, “Multiple object tracking via feature pyramid siamese networks,” IEEE Access, vol. 7, pp. 8181–8194, 2019

    2019

  52. [61]

    Joint cost minimization for multi-object tracking,

    A. Boragule and M. Jeon, “Joint cost minimization for multi-object tracking,” in 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) , Aug 2017, pp. 1–6

    2017

  53. [62]

    Simple online and realtime tracking,

    A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, “Simple online and realtime tracking,” in 2016 IEEE International Conference on Image Processing (ICIP), Sep. 2016, pp. 3464–3468

    2016

  54. [63]

    Tracking the untrackable: Learning to track multiple cues with long-term dependencies,

    A. Sadeghian, A. Alahi, and S. Savarese, “Tracking the untrackable: Learning to track multiple cues with long-term dependencies,” in 2017 15 IEEE International Conference on Computer Vision (ICCV) , 2017, pp. 300–311

    2017

  55. [64]

    Real-time online multi-object tracking in compressed domain,

    Q. Liu, B. Liu, Y . Wu, W. Li, and N. Yu, “Real-time online multi-object tracking in compressed domain,”IEEE Access, vol. 7, pp. 76 489–76 499, 2019

    2019

  56. [65]

    Tracking multiple persons based on a variational bayesian model,

    Y . Ban, S. Ba, X. Alameda-Pineda, and R. Horaud, “Tracking multiple persons based on a variational bayesian model,” in Computer Vision – ECCV 2016 Workshops , G. Hua and H. J ´egou, Eds. Cham: Springer International Publishing, 2016, pp. 52–67

    2016

  57. [66]

    Arcface: Additive angular margin loss for deep face recognition,

    J. Deng, J. Guo, N. Xue, and S. Zafeiriou, “Arcface: Additive angular margin loss for deep face recognition,” in The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , June 2019

    2019

  58. [67]

    Online multi- object tracking with instance-aware tracker and dynamic model refreshment,

    P. Chu, H. Fan, C. C. Tan, and H. Ling, “Online multi- object tracking with instance-aware tracker and dynamic model refreshment,” CoRR, vol. abs/1902.08231, 2019. [Online]. Available: http://arxiv.org/abs/1902.08231

    1902 arXiv

  59. [70]

    Focal loss for dense object detection,

    T. Lin, P. Goyal, R. B. Girshick, K. He, and P. Doll ´ar, “Focal loss for dense object detection,” CoRR, vol. abs/1708.02002, 2017. [Online]. Available: http://arxiv.org/abs/1708.02002

    2017 arXiv

  60. [2017]

    Available: http://arxiv.org/abs/1711.08565

    [Online]. Available: http://arxiv.org/abs/1711.08565

    arXiv

  61. [2019]

    Available: http://www.sciencedirect.com/science/article/ pii/S0273117719302728

    [Online]. Available: http://www.sciencedirect.com/science/article/ pii/S0273117719302728

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.