Pith. sign in

REVIEW 4 major objections 6 minor 74 references

High Performance Visual Object Tracking with Unified Convolutional Networks

T0 review · 4 major / 6 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 Training features and tracking together as one convolution produces top-ranked accuracy at 58 frames per second.

desk verdict Competent real-time tracker paper with strong ablations, but the headline VOT ranking rests on an unverified comparability assumption. read the letter →

arxiv 1908.09445 v1 pith:3MHCVLMC submitted 2019-08-26 cs.RO cs.CVeess.IV

Zheng Zhu , Wei Zou , Guan Huang , Dalong Du , Chang Huang This is my paper · ORCID
classification cs.ROcs.CVeess.IV
keywords visualobjecttrackingfullyconvolutionalnetworkend-to-endtrainingcorrelationfilterpeak-versus-noiseratioreal-timescaleestimationunifiedtracker
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 a visual tracker can reach the accuracy of the best deep-learning trackers while running faster than real time, if the feature extractor and the tracking operation are trained together as one network. Its proposed unified convolutional tracker (UCT) treats both feature extraction and tracking as convolution operations, so the full system can be trained end-to-end. On OTB2013, OTB2015, VOT2015, and VOT2016 the paper reports top-tier accuracy, including a reported first-place expected average overlap on VOT2016 at 58 frames per second. If true, this matters because the most accurate earlier deep trackers ran at around one frame per second, which made them impractical for robots, cars, and other real-time systems.

What carries the argument

The carrying mechanism is the unified convolutional tracker (UCT): a fully convolutional network in which the feature extractor and the tracking filter are both convolution operations trained jointly. The filter is learned by minimizing the $L_2$ distance between its response map $R(x_k)$ and a Gaussian label $y_k$, with a weight-decay term, using gradient descent; this keeps the filtering step differentiable, so gradients flow back into the feature extractor. During online tracking the whole patch is scored in one forward pass, model updates are gated by a peak-versus-noise ratio $\mathrm{PNR} = (R_{\max}-R_{\min})/\mathrm{mean}(R\setminus R_{\max})$ against historical thresholds, and a separate one-dimensional convolutional filter branch estimates scale changes.

What would settle it

Re-run UCT with the official VOT2016 evaluation toolkit and the same initializations and compare the resulting EAO to the reported 0.342; if the official number is materially lower, the ranking claim fails. A second, more targeted test: keep everything identical but freeze the pretrained backbone during offline training, so only the two tracking convolution layers learn; if this frozen version matches UCT's accuracy, the paper's central claim that joint feature learning delivers the performance is not supported.

Watch

Extended reading notes

Core claim

The paper's central claim is that the usual separation in deep trackers—a feature extractor pretrained for another task and a separate tracking module—is a fixable cause of suboptimal performance. UCT replaces that separation with a single fully convolutional network in which the tracking filter is a convolution layer and the feature extractor is a convolutional network, both optimized together by gradient descent on an $L_2$ loss between the response map and a Gaussian label. Online, one forward pass produces the whole foreground response map, an update is triggered only when a peak-versus-noise ratio and the peak value stay above historical thresholds, and a one-dimensional scale filter handles size changes. The paper reports that this system scores an AUC of 0.693 on OTB2013 and 0.670 on OTB2015, and an EAO of 0.3576 and 0.342 on VOT2015 and VOT2016, ranking first on both VOT challenges' EAO lists at 58 FPS, while a lighter version runs at 154 FPS with a smaller accuracy margin.

Load-bearing premise

The whole comparison rests on the assumption that benchmark scores quoted from other trackers were produced under the same evaluation conditions as UCT's own runs, so the rankings the paper draws from them are fair.

Editorial extensions

If this is right

  • Accuracy and speed need not be opposed: a single forward pass through one network can produce both a full response map and a scale estimate, so a tracker can hold its own against much slower champions.
  • The feature extractor's learned features become tailored to tracking in general rather than inherited from a classification task, which is why the paper attributes strong results on rotation and deformation attributes to end-to-end training.
  • The peak-versus-noise-ratio update rule means the model tends to skip learning from occluded or low-confidence frames, which should reduce drift and also saves computation by updating infrequently.
  • Because UCT-lite keeps most of the accuracy at 154 FPS, the same architecture can be traded off along a speed-accuracy curve to fit different hardware constraints.
  • A tracker that runs beyond real time while keeping top accuracy makes deep tracking practical for robots, automated driving, and other latency-sensitive applications.

Reading between the lines

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

  • My inference: beyond the paper, the idea of treating the tracking filter as a trainable convolution suggests a unifying view in which Siamese trackers and correlation-filter trackers differ mainly in when and how the filter layer is updated.
  • My inference: the peak-versus-noise-ratio criterion is a generic confidence measure that could be lifted into other online-learning systems, such as detectors or re-identifiers, whenever model updates risk being poisoned by bad frames.
  • My inference: the paper reports one-step SGD updates when confident frames occur; varying the number of update steps per confident frame is a natural speed-accuracy trade-off the paper leaves untested.
  • My inference: the evaluations use benchmarks from 2015-2016, so whether the advantage persists on later and harder tracking benchmarks is an open question the paper does not address.
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

4 major / 6 minor

Summary. The paper proposes a unified convolutional tracker (UCT) that represents both feature extraction and the correlation-filter tracking operation as convolutional layers in a single fully convolutional network. The filter is learned by minimizing an L2 ridge-regression loss (Eq. 2) via gradient descent, which permits end-to-end offline training on ImageNet VID and first-frame adaptation. Online, a peak-versus-noise ratio (PNR, Eq. 4) gates model updates, and a 1-D scale-filter branch handles scale changes. The authors evaluate UCT and a lighter variant, UCT-lite, on OTB2013, OTB2015, VOT2015, and VOT2016, and report leading performance among the compared trackers at 58 FPS and 154 FPS respectively. The central claim is that UCT achieves leading accuracy while running beyond real-time speed.

Significance. The paper offers a clean, unified formulation: converting the correlation-filter ridge regression of Eq. (2) into a differentiable convolution layer and training it jointly with a feature extractor is a sensible and timely idea, and the ablations in Table II are internally consistent (each component improves AUC). The PNR-based adaptive update is a simple and practical mechanism, and the reported speeds (58 FPS for UCT, 154 FPS for UCT-lite) would be valuable for real-time applications. If the VOT protocol issues are resolved, the work could serve as a strong real-time baseline. However, the paper currently does not provide code, per-sequence results, or uncertainty estimates, so the empirical contribution is not yet fully verifiable.

major comments (4)
  1. [Section V-D, Section V-E, Figs. 8 and 10, Tables IV and V] The claim that 'UCT ranks 1st in 70 trackers according to EAO criterion' on VOT2016 (Section V-E) and 'ranks 1st in 61 trackers' on VOT2015 (Section V-D) is constructed by inserting the authors' own EAO numbers into the official challenge leaderboards. The manuscript does not state that UCT was run through the official VOT evaluation toolkit; that is, the reset-after-failure protocol, the exact overlap computation, the sequence-length weighting, and the VOT release used are not specified. The single sentence 'All the tracking results use the reported results to ensure a fair comparison' (Section V) addresses the provenance of the competing trackers' numbers, not the protocol used for UCT's own numbers. Since the VOT2016 margin over CCOT is 0.342 vs 0.331 (about 3.2% relative), a modest difference in evaluation protocol could alter the ranking. Please provide the exact VOT version, confirm use of the official evaluation scripts, and release per-sequence overlap curves and raw results.
  2. [Section V-D] The exclusion of MDNet from the VOT2015 comparison relies on the statement that 'MDNet [50] is not compatible with the latest VOT rules because of OTB training data.' However, the VOT2016 results in Figure 10 and Table V include the entry 'MDNet_N,' whose relationship to reference [50] is not clarified in the paper. The paper does not explain why OTB-trained MDNet is ineligible for VOT2015 but appears in VOT2016. If the eligibility rules differ between the two challenges, this should be stated explicitly; as written, the VOT2015 'ranks 1st' claim is not established against the full participant field that would be expected under a consistent set of rules.
  3. [Sections V-B through V-E] The performance differences that support the 'leading performance' claims are reported without any measure of uncertainty. For instance, on OTB2013 the success AUC advantage over BACF is 0.693 vs 0.657 (a 5.5% relative difference), on OTB2015 the precision advantage over HDT is 0.899 vs 0.848, and on VOT2016 the EAO advantage over CCOT is 0.342 vs 0.331. These margins are small relative to the known variance of tracking benchmarks, yet no per-sequence results, confidence intervals, or paired statistical tests are given. The paper should report per-sequence overlap values and perform appropriate significance testing before concluding that UCT 'outperforms all the other trackers' (Section V-C).
  4. [Abstract, Section V-B, Section V-C] The abstract claims 'leading performance on these benchmarks' without qualification, but on OTB2013 (Section V-B) and OTB2015 (Section V-C) the comparison set is restricted to a selected list of real-time and near-real-time trackers; offline top performers such as MDNet, CCOT, and DeepSRDCF are not included in the OTB plots. Thus the OTB results demonstrate leadership within a subset, not across the full state of the art. The claim should be narrowed to 'leading performance among real-time trackers on OTB' or the OTB experiments should be extended to include the broader set of top-performing trackers.
minor comments (6)
  1. [Section III-B, Eq. (4)] Please define the set R\Rmax explicitly; in particular, state whether all entries equal to Rmax are removed and whether the mean is over the remaining spatial responses.
  2. [Section IV-A, Eq. (6) and Algorithm 1] Clarify whether the thresholds in Eq. (6) are accumulated over all frames up to the current frame or only over frames where the model was updated, and whether the current frame's PNR and Rmax are included in the threshold before the comparison. The current pseudocode order makes this ambiguous.
  3. [Section V-E, Figure 10] The text says UCT 'ranks 1st in 70 trackers,' but the horizontal axis in Figure 10 extends to 71; please reconcile the reported number of trackers.
  4. [Section V] The sentence 'All the tracking results use the reported results to ensure a fair comparison' should be rewritten to specify which numbers are taken from prior publications and which are measured by the authors, and to describe the evaluation protocol used for the authors' own numbers.
  5. [Section V-A] The paper states that code and results 'will be made publicly available,' but no link is provided; for a system paper, including code and per-sequence raw results at submission is important for reproducibility.
  6. [Section IV-B, Eq. (7)] The scale branch is described as 'inspired by [29]' but the details of how the 1-D scale filter is trained and how its output is combined with the translation response are not fully specified; please add the training objective and the rule for updating the scale model.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: UCT's central claims are empirical benchmark results, not derivations from their own inputs; the only self-citation is a normal extension note and is not load-bearing.

full rationale

The paper's central claims are empirical: UCT achieves leading performance on OTB2013, OTB2015, VOT2015, and VOT2016 while running in real time. There is no derivation chain in which a predicted quantity is equivalent by construction to a fitted input. The PNR-based model update uses an online adaptive threshold computed from the historical average of PNR and Rmax (Eq. 4-6); it is a decision rule, not a parameter fitted to benchmark outcomes, so it does not constitute a fitted input called a prediction. The scale branch is a separate 1D convolutional filter trained on first-frame samples, following standard scale-estimation practice. The paper's self-citation to its ICCVW 2017 work [42] is used to state that the present paper extends that work, but the current paper changes the backbone, training data, and training strategy and reports new experiments on four benchmarks; the performance claims rest on those new measurements, not on the cited paper. The VOT rankings are constructed by comparing self-reported EAO numbers with reported results from challenge participants, and the paper states it follows the latest VOT rules; any concern about protocol comparability is a correctness or verification issue, not a circular reduction. No step in the paper reduces to its own inputs, and no load-bearing claim is justified solely by a self-citation. Therefore the appropriate finding is no significant circularity.

Assumptions & free parameters 10 free parameters · 4 assumptions · 1 invented entities

The paper introduces no new physical entities, but it rests on several training and domain assumptions: ImageNet VID transfer, the convexity of the filter subproblem, and a heuristic PNR update rule. The hyperparameters are numerous and hand-set; none are themselves fitted to the test benchmarks, but the overall system is tuned on the evaluation datasets without a clear validation/test split.

free parameters (10)
  • Offline learning rate = 1e-5
    Used for 30 epochs of offline fine-tuning (Section V-A). Hand-set, not justified by analysis.
  • First-frame learning rate = 5e-7
    Used for 50 SGD steps on the first frame (Section V-A). Hand-set.
  • Online update learning rate = 1e-7
    Used for one-step SGD model updates (Section V-A). Hand-set.
  • Weight decay lambda offline = 0.005
    Regularization in equation (2) during offline training (Section V-A).
  • Weight decay lambda first frame = 0.01
    Regularization during first-frame adaptation (Section V-A).
  • Scale filter size S = 33
    Number of scale filters in equation (7) (Section V-A).
  • Scale factor a = 1.02
    Geometric scale step in equation (7) (Section V-A).
  • Jittering translation and scale = 0.05 and 0.02
    Data augmentation for offline training (Section V-A).
  • Hann window padding = 1.8
    Cropping padding relative to target size (Section III-A, Figure 1 caption).
  • Fine-tuned layer counts = last 6 of VGG, last 3 of ZF
    Which convolutional layers are updated in offline training (Section III-C). Design choice.
assumptions (4)
  • domain assumption ImageNet VID training data is representative of generic visual tracking targets
    The offline training uses VID train and validation sets (Section V-A); if the training distribution does not transfer to the OTB/VOT test videos, the reported performance would not generalize. No domain-adaptation analysis is provided.
  • standard math Convexity of the L2 loss in equation (2) allows gradient descent to converge to a good optimum
    Section III-C states the objective is convex and that limited gradient steps give approximate global optima. This convexity holds only for the filter subproblem with fixed features, not for the joint end-to-end training that updates the backbone network.
  • ad hoc to paper The PNR statistic is a reliable indicator of tracking confidence
    Equation (4) defines PNR, and the update rule in Section IV-A assumes that high PNR jointly with high Rmax means the current estimate is trustworthy. No calibration or analysis is given; it is a heuristic specific to this paper.
  • domain assumption The scale branch from fDSST transfers to the convolutional framework
    Section IV-B adopts the 1D scale filter idea from fDSST without re-deriving it; the paper assumes the original scale estimation behavior carries over to the new feature representation.
invented entities (1)
  • Peak-versus-noise ratio (PNR)
    purpose: A new response-map statistic used to decide when to update the tracking model online.
    The paper introduces PNR as a new criterion (equation 4) and demonstrates its usefulness only through the paper's own benchmark experiments and ablations. No external validation or theoretical justification is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of High Performance Visual Object Tracking with Unified Convolutional Networks." pith.science (2026). https://pith.science/paper/3MHCVLMC

@misc{pith2026190809445,
  author       = {Pith},
  title        = {Pith review of: High Performance Visual Object Tracking with Unified Convolutional Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3MHCVLMC}},
  note         = {Machine review of arXiv:1908.09445}
}
read the original abstract

Convolutional neural networks (CNN) based tracking approaches have shown favorable performance in recent benchmarks. Nonetheless, the chosen CNN features are always pre-trained in different tasks and individual components in tracking systems are learned separately, thus the achieved tracking performance may be suboptimal. Besides, most of these trackers are not designed towards real-time applications because of their time-consuming feature extraction and complex optimization details. In this paper, we propose an end-to-end framework to learn the convolutional features and perform the tracking process simultaneously, namely, a unified convolutional tracker (UCT). Specifically, the UCT treats feature extractor and tracking process both as convolution operation and trains them jointly, which enables learned CNN features are tightly coupled with tracking process. During online tracking, an efficient model updating method is proposed by introducing peak-versus-noise ratio (PNR) criterion, and scale changes are handled efficiently by incorporating a scale branch into network. Experiments are performed on four challenging tracking datasets: OTB2013, OTB2015, VOT2015 and VOT2016. Our method achieves leading performance on these benchmarks while maintaining beyond real-time speed.

Figures

Figures reproduced from arXiv: 1908.09445 by the authors.

Figure 1
Figure 1. The overall UCT architecture. In each frame, patch is cropped around ground truth and resized into [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. From left to right: images, response maps without off [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Updating results of UCT and UCT No PNR (UCT without PNR criterion). The first row shows frames that the target is occluded by distractor. The second row is corre￾sponding response maps. Rmax still keeps large in occlusion while PNR significantly decreases. So the unwanted updating is avoided by considering PNR constraint simultaneously. The red and blue boxes in last image are tracking results of UCT and UCT No PNR,… view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Precision and success plots on OTB2013. The numbers [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Attributes performance on OTB2013. Best viewed on color display. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Precision and success plots on OTB2015. The numbers [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Attributes performance on OTB2015. Best viewed on color display. [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 10
Figure 10. Figure 10: EAO ranking with trackers in VOT2016. The better [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 9
Figure 9. Figure 9: EAO scores for each attribute on the VOT2015 dataset. [PITH_FULL_IMAGE:figures/full_fig_p009_9.png]
Figure 11
Figure 11. Figure 11: EAO scores for each attribute on the VOT2016 dataset. [PITH_FULL_IMAGE:figures/full_fig_p010_11.png]
Figure 12
Figure 12. Figure 12: Performance and speed of our tracker and some state [PITH_FULL_IMAGE:figures/full_fig_p010_12.png]
Figure 13
Figure 13. Figure 13: Comparisons of our approach with four state-of-the-art trackers in the changing scenario [PITH_FULL_IMAGE:figures/full_fig_p011_13.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

74 extracted references · 66 canonical work pages

  1. [50]

    Learning multi-domain convolu- tional neural networks for visual tracking,

    H. Nam and B. Han, “Learning multi-domain convolu- tional neural networks for visual tracking,” in Proceed- ings of the IEEE Conference on Computer Vision and Pattern Recognition, June 2016

    work page 2016

  2. [1]

    Sparse representation for crowd attributes recognition,

    A. N. Shuaibu, I. Faye, Y . S. Ali, N. Kamel, M. N. Saad, and A. S. Malik, “Sparse representation for crowd attributes recognition,” IEEE Access , vol. 5, pp. 10 422– 10 433, 2017

    work page 2017

  3. [2]

    Research on automatic parking systems based on parking scene recognition,

    S. Ma, H. Jiang, M. Han, J. Xie, and C. Li, “Research on automatic parking systems based on parking scene recognition,” IEEE Access , vol. 5, pp. 21 901–21 917, 2017

    work page 2017

  4. [3]

    Fastpose: Towards real-time pose estimation and tracking via scale-normalized multi-task networks,

    J. Zhang, Z. Zhu, W. Zou, P. Li, Y . Li, H. Su, and G. Huang, “Fastpose: Towards real-time pose estimation and tracking via scale-normalized multi-task networks,” arXiv preprint arXiv:1908.05593 , 2019

    arXiv 1908

  5. [4]

    Exploiting Offset-guided Network for Pose Estimation and Tracking

    R. Zhang, Z. Zhu, P. Li, R. Wu, C. Guo, G. Huang, and H. Xia, “Exploiting offset-guided network for pose esti- mation and tracking,” arXiv preprint arXiv:1906.01344 , 2019

    work page Pith review arXiv 1906

  6. [5]

    State-aware re-identification feature for multi-target multi-camera tracking,

    P. Li, J. Zhang, Z. Zhu, Y . Li, L. Jiang, and G. Huang, “State-aware re-identification feature for multi-target multi-camera tracking,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2019, pp. 0–0

    work page 2019

  7. [6]

    Selection of observation position and orientation in visual servoing with eye-in-vehicle configuration for manipulator,

    H.-X. Ma, W. Zou, Z. Zhu, C. Zhang, and Z.-B. Kang, “Selection of observation position and orientation in visual servoing with eye-in-vehicle configuration for manipulator,” International Journal of Automation and Computing, pp. 1–14

    work page

  8. [7]

    Adaptive tra- jectory tracking of wheeled mobile robots based on a fish-eye camera,

    Z. Kang, W. Zou, H. Ma, and Z. Zhu, “Adaptive tra- jectory tracking of wheeled mobile robots based on a fish-eye camera,” International Journal of Control, Automation and Systems , pp. 1–13

    work page

Show all 74 references
  1. [8]

    A velocity compensation visual servo method for oculomotor con- trol of bionic eyes,

    Z. Zhu, W. Zou, Q. Wang, F. Zhang et al. , “A velocity compensation visual servo method for oculomotor con- trol of bionic eyes,” 2018

    2018

  2. [9]

    Motion control in saccade and smooth pursuit for bionic eye based on three- dimensional coordinates,

    Q. Wang, W. Zou, D. Xu, and Z. Zhu, “Motion control in saccade and smooth pursuit for bionic eye based on three- dimensional coordinates,” Journal of Bionic Engineering, vol. 14, no. 2, pp. 336–347, 2017

    2017

  3. [10]

    Optical flow based real-time moving object detection in unconstrained scenes,

    J. Huang, W. Zou, J. Zhu, and Z. Zhu, “Optical flow based real-time moving object detection in unconstrained scenes,” arXiv preprint arXiv:1807.04890 , 2018

    2018 arXiv

  4. [11]

    Optical flow based online moving foreground analysis,

    J. Huang, W. Zou, and Z. Zhu, “Optical flow based online moving foreground analysis,” arXiv preprint arXiv:1811.07256, 2018. 11 UCT(Ours) SiamFCCCOT CFNet PTAV Fig. 13: Comparisons of our approach with four state-of-the-art trackers in the changing scenario skiing, singer2, iro...

    2018 arXiv

  5. [12]

    An effi- cient optical flow based motion detection method for non-stationary scenes,

    J. Huang, W. Zou, Z. Zhu, and J. Zhu, “An effi- cient optical flow based motion detection method for non-stationary scenes,” arXiv preprint arXiv:1811.08290, 2018

    2018 arXiv

  6. [13]

    Motion cue based instance-level moving object detection,

    J. Huang, W. Zou, Z. Zhu, J. Zhu et al. , “Motion cue based instance-level moving object detection,” 2019

    2019

  7. [14]

    Std: A stereo tracking dataset for evaluating binocular tracking algorithms,

    Z. Zhu, W. Zou, Q. Wang, and F. Zhang, “Std: A stereo tracking dataset for evaluating binocular tracking algorithms,” in 2016 IEEE International Conference on Robotics and Biomimetics (ROBIO) . IEEE, 2016, pp. 2215–2220

    2016

  8. [15]

    Object tracking benchmark,

    Y . Wu, J. Lim, and M. H. Yang, “Object tracking benchmark,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 37, no. 9, pp. 1834–1848, 2015

    2015

  9. [16]

    Visual tracking: An exper- imental survey,

    A. W. Smeulders, D. M. Chu, R. Cucchiara, S. Calderara, A. Dehghan, and M. Shah, “Visual tracking: An exper- imental survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 36, no. 7, pp. 1442–1468, 2014

    2014

  10. [17]

    The sixth visual object tracking vot2018 challenge results,

    M. Kristan, A. Leonardis, J. Matas, M. Felsberg, R. Pflugfelder, L. ˇC. Zajc, T. V oj´ır, G. Bhat, A. Lukeˇziˇc, A. Eldesokey et al. , “The sixth visual object tracking vot2018 challenge results,” in European Conference on Computer Vision. Springer, Cham, 2018, pp. 3–53

    2018

  11. [18]

    The visual object tracking vot2015 challenge results,

    K. Matej, J. Matas, A. Leonardis, M. Felsberg, L. ˇCehovin, G. Fern ´andez, T. V oj´ıˇr, G. H ¨ager, G. Nebe- hay, R. Pflugfelder et al. , “The visual object tracking vot2015 challenge results,” in Workshop on the Visual Object Tracking Challenge (VOT, in conjunction with 12 IC...

    2015

  12. [19]

    Online object tracking with sparse prototypes,

    D. Wang, H. Lu, and M.-H. Yang, “Online object tracking with sparse prototypes,” IEEE Transactions on Image Processing, vol. 22, no. 1, pp. 314–325, 2013

    2013

  13. [20]

    Inverse sparse tracker with a locally weighted distance metric,

    D. Wang, H. Lu, Z. Xiao, and M.-H. Yang, “Inverse sparse tracker with a locally weighted distance metric,” IEEE Transactions on Image Processing , vol. 24, no. 9, pp. 2646–2657, 2015

    2015

  14. [21]

    Struck: Structured output tracking with kernels,

    S. Hare, S. Golodetz, A. Saffari, V . Vineet, M.-M. Cheng, S. L. Hicks, and P. H. Torr, “Struck: Structured output tracking with kernels,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 38, no. 10, pp. 2096–2109, 2016

    2016

  15. [22]

    Real-time tracking via on-line boosting,

    H. Grabner, M. Grabner, and H. Bischof, “Real-time tracking via on-line boosting,” in Proceedings of the British Machine Vision Conference , 2006

    2006

  16. [23]

    Robust object tracking with online multiple instance learning,

    B. Babenko, M.-H. Yang, and S. Belongie, “Robust object tracking with online multiple instance learning,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 33, no. 8, pp. 1619–1632, 2011

    2011

  17. [24]

    High-speed tracking with kernelized correlation filters,

    J. F. Henriques, R. Caseiro, P. Martins, and J. Batista, “High-speed tracking with kernelized correlation filters,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 37, no. 3, p. 583, 2015

    2015

  18. [25]

    Learning spatially regularized correlation filters for vi- sual tracking,

    M. Danelljan, G. Hager, F. S. Khan, and M. Felsberg, “Learning spatially regularized correlation filters for vi- sual tracking,” in Proceedings of the IEEE International Conference on Computer Vision , 2015, pp. 4310–4318

    2015

  19. [26]

    Long- term correlation tracking,

    C. Ma, X. Yang, C. Zhang, and M.-H. Yang, “Long- term correlation tracking,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 5388–5396

    2015

  20. [27]

    A scale adaptive kernel correlation filter tracker with feature integration,

    Y . Li and J. Zhu, “A scale adaptive kernel correlation filter tracker with feature integration,” in Proceedings of the European Conference on Computer Vision Workshop , 2014, pp. 254–265

    2014

  21. [28]

    Ex- ploiting the circulant structure of tracking-by-detection with kernels,

    J. F. Henriques, C. Rui, P. Martins, and J. Batista, “Ex- ploiting the circulant structure of tracking-by-detection with kernels,” in Proceedings of the European Confer- ence on Computer Vision , 2012, pp. 702–715

    2012

  22. [29]

    Discriminative scale space tracking,

    M. Danelljan, G. H ¨ager, F. S. Khan, and M. Felsberg, “Discriminative scale space tracking,”IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 39, no. 8, pp. 1561–1575, 2017

    2017

  23. [30]

    Learning background-aware correlation filters for visual tracking,

    H. Kiani Galoogahi, A. Fagg, and S. Lucey, “Learning background-aware correlation filters for visual tracking,” in Proceedings of the IEEE International Conference on Computer Vision, Oct 2017

    2017

  24. [31]

    Imagenet classification with deep convolutional neural networks,

    A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” in Proceedings of the Advances in Neural Information Processing Systems, 2012, pp. 1097–1105

    2012

  25. [32]

    Deep resid- ual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep resid- ual learning for image recognition,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 770–778

    2016

  26. [33]

    Faster r-cnn: towards real-time object detection with region proposal networks,

    S. Ren, K. He, R. Girshick, and J. Sun, “Faster r-cnn: towards real-time object detection with region proposal networks,” in Proceedings of the Advances in Neural Information Processing Systems , 2015, pp. 91–99

    2015

  27. [34]

    Fully convolu- tional networks for semantic segmentation,

    J. Long, E. Shelhamer, and T. Darrell, “Fully convolu- tional networks for semantic segmentation,” in Proceed- ings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 3431–3440

    2015

  28. [35]

    Hi- erarchical convolutional features for visual tracking,

    C. Ma, J.-B. Huang, X. Yang, and M.-H. Yang, “Hi- erarchical convolutional features for visual tracking,” in Proceedings of the IEEE International Conference on Computer Vision, December 2015

    2015

  29. [36]

    Hedged deep tracking,

    Y . Qi, S. Zhang, L. Qin, H. Yao, Q. Huang, J. Lim, and M.-H. Yang, “Hedged deep tracking,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June 2016

    2016

  30. [37]

    Beyond correlation filters: Learning continuous convo- lution operators for visual tracking,

    M. Danelljan, A. Robinson, F. S. Khan, and M. Felsberg, “Beyond correlation filters: Learning continuous convo- lution operators for visual tracking,” in Proceedings of the European Conference on Computer Vision , 2016, pp. 472–488

    2016

  31. [38]

    Convolutional features for correlation filter based visual tracking,

    M. Danelljan, G. Hager, F. S. Khan, and M. Felsberg, “Convolutional features for correlation filter based visual tracking,” in Proceedings of the IEEE International Con- ference on Computer Vision Workshop , 2015, pp. 621– 629

    2015

  32. [39]

    Fully-convolutional siamese networks for object tracking,

    L. Bertinetto, J. Valmadre, J. F. Henriques, A. Vedaldi, and P. H. S. Torr, “Fully-convolutional siamese networks for object tracking,” in Proceedings of the European Conference on Computer Vision Workshop , 2016, pp. 850–865

    2016

  33. [40]

    Visual track- ing with fully convolutional networks,

    L. Wang, W. Ouyang, X. Wang, and H. Lu, “Visual track- ing with fully convolutional networks,” in Proceedings of the IEEE International Conference on Computer Vision , 2015, pp. 3119–3127

    2015

  34. [41]

    The visual object tracking vot2016 challenge results,

    M. Kristan, A. Leonardis, J. Matas, M. Felsberg, R. Pflugfelder, L. ehovin, T. V ojr, G. Hger, A. Lukei, and G. Fernndez, “The visual object tracking vot2016 challenge results,” in Proceedings of the European Con- ference on Computer Vision Workshop . Springer Inter- national P...

    2016

  35. [42]

    UCT: Learning unified convolutional networks for real-time visual tracking,

    Z. Zhu, G. Huang, W. Zou, D. Du, and C. Huang, “UCT: Learning unified convolutional networks for real-time visual tracking,” in 2017 IEEE International Conference on Computer Vision Workshops (ICCVW) , Oct 2017, pp. 1973–1982

    2017

  36. [43]

    Two-stream gated fusion convnets for action recognition,

    J. Zhu, W. Zou, and Z. Zhu, “Two-stream gated fusion convnets for action recognition,” in 2018 24th Interna- tional Conference on Pattern Recognition (ICPR). IEEE, 2018, pp. 597–602

    2018

  37. [44]

    Attention-guided unified network for panoptic segmentation,

    Y . Li, X. Chen, Z. Zhu, L. Xie, G. Huang, D. Du, and X. Wang, “Attention-guided unified network for panoptic segmentation,” in CVPR, 2019

    2019

  38. [45]

    Action machine: Rethinking action recognition in trimmed videos,

    J. Zhu, W. Zou, L. Xu, Y . Hu, Z. Zhu, M. Chang, J. Huang, G. Huang, and D. Du, “Action machine: Rethinking action recognition in trimmed videos,” arXiv preprint arXiv:1812.05770, 2018

    2018 arXiv

  39. [46]

    Learning gating convnet for two-stream based methods in action recognition,

    J. Zhu, W. Zou, and Z. Zhu, “Learning gating convnet for two-stream based methods in action recognition,” arXiv preprint arXiv:1709.03655, vol. 1, no. 2, p. 6, 2017. 13

    2017 arXiv

  40. [47]

    Transferring rich feature hierarchies for robust visual tracking,

    N. Wang, S. Li, A. Gupta, and D.-Y . Yeung, “Transferring rich feature hierarchies for robust visual tracking,” arXiv preprint arXiv:1501.04587, 2015

    2015 arXiv

  41. [48]

    Deeptrack: Learning dis- criminative feature representations online for robust vi- sual tracking,

    H. Li, Y . Li, and F. Porikli, “Deeptrack: Learning dis- criminative feature representations online for robust vi- sual tracking,” IEEE Transactions on Image Processing , vol. 25, no. 4, pp. 1834–1848, 2016

    2016

  42. [49]

    Multi-hierarchical independent correlation filters for vi- sual tracking,

    S. Bai, Z. He, T.-B. Xu, Z. Zhu, Y . Dong, and H. Bai, “Multi-hierarchical independent correlation filters for vi- sual tracking,” arXiv preprint arXiv:1811.10302 , 2018

    2018 arXiv

  43. [51]

    Template matching using fast normalized cross correlation,

    K. Briechle and U. D. Hanebeck, “Template matching using fast normalized cross correlation,” in Optical Pat- tern Recognition XII , vol. 4387, 2001, pp. 95–103

    2001

  44. [52]

    Real-time track- ing of non-rigid objects using mean shift,

    D. Comaniciu, V . Ramesh, and P. Meer, “Real-time track- ing of non-rigid objects using mean shift,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition, vol. 2, 2000, pp. 142–149

    2000

  45. [53]

    Visual object tracking using adaptive correlation filters,

    D. S. Bolme, J. R. Beveridge, B. A. Draper, and Y . M. Lui, “Visual object tracking using adaptive correlation filters,” in Proceedings of the IEEE Conference on Com- puter Vision and Pattern Recognition , 2010, pp. 2544– 2550

    2010

  46. [54]

    Staple: Complementary learners for real- time tracking,

    L. Bertinetto, J. Valmadre, S. Golodetz, O. Miksik, and P. H. S. Torr, “Staple: Complementary learners for real- time tracking,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , June 2016

    2016

  47. [55]

    Correla- tion filters with limited boundaries,

    H. Kiani Galoogahi, T. Sim, and S. Lucey, “Correla- tion filters with limited boundaries,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 4630–4638

    2015

  48. [56]

    Learning to track at 100 fps with deep regression networks,

    D. Held, S. Thrun, and S. Savarese, “Learning to track at 100 fps with deep regression networks,” in Proceedings of the European Conference on Computer Vision , 2016, pp. 749–765

    2016

  49. [57]

    End-to-end representation learning for correlation filter based tracking,

    J. Valmadre, L. Bertinetto, J. F. Henriques, A. Vedaldi, and P. H. S. Torr, “End-to-end representation learning for correlation filter based tracking,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017

    2017

  50. [58]

    Dcfnet: Discriminant correlation filters network for visual track- ing,

    Q. Wang, J. Gao, J. Xing, M. Zhang, and W. Hu, “Dcfnet: Discriminant correlation filters network for visual track- ing,” arXiv preprint arXiv:1704.04057 , 2017

    2017 arXiv

  51. [59]

    End-to-end flow correlation tracking with spatial-temporal attention,

    Z. Zhu, W. Wu, W. Zou, and J. Yan, “End-to-end flow correlation tracking with spatial-temporal attention,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2018

    2018

  52. [60]

    End-to-end video-level representation learning for action recognition,

    J. Zhu, Z. Zhu, and W. Zou, “End-to-end video-level representation learning for action recognition,” in 2018 24th International Conference on Pattern Recognition (ICPR). IEEE, 2018, pp. 645–650

    2018

  53. [61]

    High performance visual tracking with siamese region proposal network,

    B. Li, J. Yan, W. Wu, Z. Zhu, and X. Hu, “High performance visual tracking with siamese region proposal network,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2018, pp. 8971–8980

    2018

  54. [62]

    Distractor-aware siamese networks for visual object tracking,

    Z. Zhu, Q. Wang, B. Li, W. Wu, J. Yan, and W. Hu, “Distractor-aware siamese networks for visual object tracking,” in Proceedings of the European Conference on Computer Vision (ECCV) , 2018, pp. 101–117

    2018

  55. [63]

    Densebox: Unifying landmark localization with end to end object detection,

    L. Huang, Y . Yang, Y . Deng, and Y . Yu, “Densebox: Unifying landmark localization with end to end object detection,” arXiv preprint arXiv:1509.04874 , 2015

    2015 arXiv

  56. [64]

    Learning a deep compact im- age representation for visual tracking,

    N. Wang and D.-Y . Yeung, “Learning a deep compact im- age representation for visual tracking,” in Proceedings of the Advances in Neural Information Processing Systems , 2013, pp. 809–817

    2013

  57. [65]

    Online object tracking: A benchmark,

    Y . Wu, J. Lim, and M. H. Yang, “Online object tracking: A benchmark,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2013, pp. 2411–2418

    2013

  58. [66]

    The visual object tracking vot2015 challenge results,

    M. Kristan, J. Matas, A. Leonardis, and M. Felsberg, “The visual object tracking vot2015 challenge results,” in Proceedings of the IEEE International Conference on Computer Vision Workshop , 2015, pp. 564–586

    2015

  59. [67]

    Very deep convolu- tional networks for large-scale image recognition,

    K. Simonyan and A. Zisserman, “Very deep convolu- tional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556, 2014

    2014 arXiv

  60. [68]

    Visualizing and under- standing convolutional networks,

    M. D. Zeiler and R. Fergus, “Visualizing and under- standing convolutional networks,” in Proceedings of the European Conference on Computer Vision . Springer, 2014, pp. 818–833

    2014

  61. [69]

    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, vol. 115, no. 3, pp. 211–252, 2015

    2015

  62. [70]

    Caffe: Convolutional architecture for fast feature embedding,

    Y . Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama, and T. Darrell, “Caffe: Convolutional architecture for fast feature embedding,” in Proceedings of the 22nd ACM international conference on Multimedia . ACM, 2014, pp. 675–678

    2014

  63. [71]

    Parallel tracking and verifying: A framework for real-time and high accuracy visual tracking,

    H. Fan and H. Ling, “Parallel tracking and verifying: A framework for real-time and high accuracy visual tracking,” in Proceedings of the IEEE International Con- ference on Computer Vision , 2017

    2017

  64. [72]

    Context-aware correlation filter tracking,

    M. Mueller, N. Smith, and B. Ghanem, “Context-aware correlation filter tracking,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 1396–1404

    2017

  65. [73]

    Discriminative correlation filter with channel and spatial reliability,

    A. Lukei, T. V oj, L. ehovin, J. Matas, and M. Kristan, “Discriminative correlation filter with channel and spatial reliability,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2017

    2017

  66. [74]

    Visual tracking using attention- modulated disintegration and integration,

    J. Choi, H. Jin Chang, J. Jeong, Y . Demiris, and J. Young Choi, “Visual tracking using attention- modulated disintegration and integration,” inProceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 4321–4330

    2016

Pith tools

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