Pith. sign in

REVIEW 4 major objections 6 minor 41 references

MOT FCG++: Enhanced Representation of Spatio-temporal Motion and Appearance Features

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read This paper claims that three small changes to how a tracker represents motion and appearance — a diagonally modulated GIoU, a confidence-weighted dynamic appearance embedding, and an averaged constant-velocity model — each improve the…

desk verdict Incremental MOT paper with a plausible weak-cue idea, but an uncontrolled baseline and inconsistent MOT20 numbers muddy the attribution. read the letter →

arxiv 2411.10028 v2 pith:VDC2U2JN submitted 2024-11-15 cs.CV cs.AI

Yanzhao Fang This is my paper · ORCID
classification cs.CVcs.AI
keywords multi-objecttrackinghierarchicalclusteringappearanceembeddingmotionfeaturesGIoUexponentialmovingaverageMOT17MOT20
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 claims that a clustering-based multi-object tracker can be improved without new detectors or new network architectures, by refining how trajectory-level motion and appearance features are represented. Starting from MOT FCG, the authors replace plain IoU with a Diagonal Modulated GIoU that encodes both position and box shape, replace the median-element appearance feature with a confidence-weighted exponential moving average over the trajectory, and replace two-frame velocity estimates with a velocity averaged over the last N frames. They report consistent gains across MOT17, MOT20, and DanceTrack, with the largest reported improvement being a +2.6 IDF1 gain on the MOT20 test set. A sympathetic reader would care because the claim is that weak cues that existing clustering trackers discard — box diagonals, detection confidence, and multi-frame motion history — are enough to make association more robust in crowded and occluded scenes.

What carries the argument

The central machinery is the trio of representation upgrades layered onto MOT FCG's hierarchical clustering: (1) DGIoU, a spatial similarity score that multiplies GIoU by the ratio of the enclosing rectangle's side lengths, making the metric sensitive to box shape and relative scale; (2) Dynamic Appearance, an exponential moving average of ReID embeddings whose update weight decreases as detection confidence drops, so low-quality frames contribute less; and (3) ACV, a constant-velocity motion model that averages displacement over the last $N$ frames to suppress observation noise. The paper uses these as weak cues within the existing clustering framework, claiming that they make the trajectory-level features more global and more robust without changing the hierarchical association structure itself.

What would settle it

Reproduce MOT FCG++ with the same fine-tuned YOLOX detector and SBS ReID features but swap DGIoU back to plain IoU; if HOTA on MOT17-val does not drop below the reported 76.14, the paper's attribution of the gain to the diagonal modulation fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that MOT FCG's two feature representations are the bottleneck: IoU misrepresents positional and shape relationships between bounding boxes, and the median-element appearance feature ignores both global trajectory information and detection quality. The proposed replacements are Diagonal Modulated GIoU, defined as $d_{\mathrm{DGIoU}} = 1 - (L_2/L_1)\cdot\mathrm{GIoU}$, where $L_1,L_2$ are the box-side lengths entering the enclosing rectangle, so that the metric reflects aspect-ratio and scale; Dynamic Appearance, an EMA with adaptive weight $\beta_t$ driven by detection confidence $s_{\mathrm{det}}$; and Average Constant Velocity Modeling, $v = (x_t - x_{t-N})/N$, which averages displacement over the last $N$ frames instead of using the last two. The paper argues each module contributes a small but consistent gain, with DGIoU helping across all three datasets and Dynamic Appearance offering the largest identity-maintenance gains during occlusion.

Load-bearing premise

The claim that the three modules cause the improvements depends on the baseline MOT FCG numbers being directly comparable to the new runs, but the new runs use a fine-tuned detector and a different ReID network than the published baseline did.

Editorial extensions

If this is right

  • If the claim holds, clustering-based trackers can be improved inexpensively by enriching the spatial metric with shape information, since DGIoU is reported to beat plain IoU and width- or height-modulated variants on MOT17.
  • The confidence-weighted dynamic appearance representation is a drop-in replacement for median-element or mean appearance features, and the paper reports it prevents identity switches during occlusion in a qualitative MOT17 example.
  • Averaging velocity over N frames stabilizes motion prediction; the paper reports N = 9 as the best setting on MOT17 and shows that the gain persists over a wide range of N.
  • The largest relative gain is on MOT20's dense, heavily occluded scenes, where IDF1 rises from 69.7 to 72.3, suggesting the improvements are especially relevant when appearance evidence is noisy.
  • With the optional StrongSORT post-processing (AFLink+GSI), the paper reports further gains to 64.1 HOTA and 79.2 MOTA on MOT17, indicating the modules compose with downstream trajectory smoothing.

Reading between the lines

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

  • A natural extension, not tested in the paper, is whether diagonal modulation generalizes to other spatial metrics such as center-distance or scaled IoU, since the success of $L_2/L_1$ suggests any cue carrying aspect-ratio information could help clustering association.
  • The reported optimum of N = 9 on MOT17 may be dataset-dependent; on DanceTrack, where motion is nonlinear and fast-changing, shorter averaging windows might be safer, and this is a directly testable hyperparameter sweep.
  • Because the paper's baseline numbers come from the published MOT FCG paper while its own runs use a fine-tuned YOLOX detector and SBS ReID features, the cleanest test of the three modules is to toggle each on and off inside a single codebase with identical detector and ReID weights; the paper's ablation does not fully isolate the modules from the component upgrades.
  • The confidence-threshold formulation of Dynamic Appearance suggests a natural robustness test: degrade detection confidence artificially and check whether the adaptive weight prevents identity switches better than a fixed-weight EMA.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes three modifications to the hierarchical clustering tracker MOT FCG: Diagonal Modulated GIoU (DGIoU) for spatial motion similarity, a dynamic confidence-weighted appearance representation, and Average Constant Velocity Modeling (ACV) for motion prediction. The authors report consistent improvements over MOT FCG on MOT17, MOT20, and DanceTrack, with headline numbers of 63.1 HOTA, 76.9 MOTA, and 78.2 IDF1 on the MOT17 test set. The paper includes benchmark comparisons and ablation studies on validation splits.

Significance. If the attribution were established, the work would offer a simple, incremental improvement to clustering-based MOT, with three small modules that are easy to understand and reproduce. The paper is honest about the modest size of the gains and validates on three benchmarks with complementary characteristics. The main strengths are the breadth of experiments and the clear separation of the three proposed components. However, the central claim that the proposed modules cause the reported gains is currently undermined by an uncontrolled baseline comparison and by a direct inconsistency between the ablation table and the main results table for MOT20.

major comments (4)
  1. [4.1, Tables 1 and 4] The baseline is not a controlled comparator. Section 4.1 states that the YOLOX detector is fine-tuned and that ReID features are extracted with the SBS network, while the MOT FCG numbers are taken from the published paper. Since detection quality and embedding quality directly affect tracklet clustering, the reported deltas (+0.5 HOTA on MOT17, +2.6 IDF1 on MOT20) cannot be attributed to DGIoU, Dynamic Appearance, and ACV. The authors should re-run MOT FCG under exactly the FCG++ detector/ReID configuration and use that as the ablation baseline, or explicitly quantify the individual contributions of the detector/ReID changes.
  2. [3.2, Eq. (1)] The normalization statement for DGIoU is mathematically incorrect. The text claims that dividing dDGIoU by 2 ensures dGIoU is in [0,1], but dDGIoU = 1 - (L2/L1)*GIoU is not guaranteed to be in [0,1] because GIoU ranges over [-1,1] and L2/L1 is an arbitrary positive ratio. If lambda_C is intended as a bounded similarity or distance, its range should be defined carefully or clipped explicitly. This is a load-bearing detail for the DGIoU contribution.
  3. [Table 4 vs. Table 3] There is a direct numerical inconsistency for MOT20. The final ablation row in Table 4 reports MOT20-val HOTA of 68.81, while Table 3 reports MOT FCG++ on the same validation split at 65.8 HOTA, and the text says the improvement is from 65.7 to 65.8. A three-point HOTA discrepancy is far larger than the claimed improvements and must be reconciled; as printed, the two tables cannot both describe the same configuration.
  4. [4.3, Figure 7, Section 4.1] The hyperparameter and ablation analysis is not fully convincing. In Section 4.3, N is selected as 9 based on HOTA on MOT17-train, but the maxima in Figure 7 differ across metrics (MOTA peaks at N=16, AssA at N=8), and no error bars or multiple-seed variance are reported. Additionally, beta_f and off are tuned per dataset on the same validation splits where the final numbers are reported, so the risk of overfitting to the benchmark is not quantified. Please report sensitivity to these choices and variance where feasible.
minor comments (6)
  1. [Figure 1 caption] The caption reports '68.1 MOTA, 72.3 IDF1 and 58.4 HOTA on the MOT17-test set', but the numbers correspond to MOT20; the dataset name is a typo.
  2. [Figure 6 captions] The two subcaptions appear garbled ('with or Dynamic Appearance Feature', 'with or Median Appearance Feature') and do not clearly distinguish the panels; 'with/without' is presumably intended.
  3. [3.3, Eqs. (2)-(3)] The notation for the EMA weight is inconsistent: Eq. (2) uses beta as a constant, while Eq. (3) defines beta_t as adaptive; please clarify the update rule and the dependence of beta_t on the detection confidence.
  4. [3.4 and 4.3] The parameter is called 'n' in the method text ('if the trajectory length is less than n') but 'N' in the ablation and equation; please use one symbol consistently.
  5. [Section 4.1] The hyperparameter names are typeset inconsistently ('beta_f and of f' instead of 'beta_f and off'); also, 'off' is not defined beyond being an offset in Eq. (1), so its role and allowed range should be stated.
  6. [References] Reference [32] is malformed ('et al Luiten, JonathonOsep'); please fix the author list and formatting.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation; benchmark gains are empirical and not forced by construction.

full rationale

This is an empirical performance-improvement paper rather than a derivation. The three proposed modules are defined by explicit formulas (DGIoU in Eq. 1, dynamic appearance in Eqs. 2-3, ACV in Eq. 4) and none of these formulas contains the reported HOTA, MOTA, or IDF1 values or the baseline metrics. No parameter is fitted to a target metric and then renamed a prediction: the per-dataset hyperparameters beta_f and off are tuned and then used in tracking, and the reported benchmark numbers are the outcome of running the tracker, not a quantity algebraically forced by the fitted values. The main weaknesses in the paper are experimental-control issues rather than circularity: the MOT FCG baseline column is taken from the original published paper rather than re-run with the fine-tuned YOLOX detector and SBS ReID features described in Section 4.1, and beta_f/off are tuned per dataset. These weaknesses weaken causal attribution of the gains to the individual modules, but they do not make the result equivalent to its inputs by construction. There are also no load-bearing self-citations or imported uniqueness theorems; the cited inspirations are external methods. Thus there is no significant circularity of the kind this axis targets.

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

The central claim rests on four free hyperparameters (beta_f, off, N, sigma), several domain assumptions about the baseline and motion model, and one ad hoc assumption that detection confidence predicts embedding quality. No new entities are introduced. The baseline MOT FCG and its assumptions are taken as given.

free parameters (4)
  • beta_f (EMA base weight) = 0.822 (MOT17), 0.66 (MOT20), 0.8 (DanceTrack)
    Sets the adaptive appearance weight in Eq. 3; tuned per dataset on validation, directly affects trajectory appearance representation.
  • off (DGIoU offset) = 0.525 (MOT17), 0.9 (MOT20), 0.1 (DanceTrack)
    Shifts the DGIoU similarity via lambda_C = min(1, dDGIoU/2 + off) in Eq. 1; no derivation, tuned per dataset.
  • N (ACV averaging window) = 9 (selected on MOT17-train)
    Number of frames averaged for constant velocity estimate in Eq. 4; chosen from ablation where MOTA peaked at N=16 but HOTA at N=9.
  • sigma (confidence threshold) = not explicitly reported; detector threshold 0.7
    Used in Eq. 3 to gate noisy detections; the threshold value is not given in the text.
assumptions (4)
  • domain assumption MOT FCG hierarchical clustering with UPGMA and median-element appearance representation is a sound baseline
    The paper builds directly on reference 24 and only replaces the median representation and velocity model; if the baseline is flawed the comparison is unfair.
  • ad hoc to paper Detection confidence correlates with appearance embedding quality
    Eq. 3 weights embeddings by s_det; no empirical evidence in this paper supports this correlation.
  • domain assumption Average velocity over N frames reduces observation noise and is suitable for the motion patterns in MOT17, MOT20, and DanceTrack
    Eq. 4 assumes zero-mean noise and roughly linear motion; DanceTrack has diverse nonlinear motion, so the assumption is questionable there.
  • ad hoc to paper GIoU and its L2/L1 diagonal modulation capture the position and shape relationship relevant for same-target association
    The metric is heuristic and validated only by ablation, without a formal argument for why the diagonal ratio is the right modulator.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MOT FCG++: Enhanced Representation of Spatio-temporal Motion and Appearance Features." pith.science (2026). https://pith.science/paper/VDC2U2JN

@misc{pith2026241110028,
  author       = {Pith},
  title        = {Pith review of: MOT FCG++: Enhanced Representation of Spatio-temporal Motion and Appearance Features},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VDC2U2JN}},
  note         = {Machine review of arXiv:2411.10028}
}
read the original abstract

The goal of multi-object tracking (MOT) is to detect and track all objects in a scene across frames, while maintaining a unique identity for each object. Most existing methods rely on the spatial-temporal motion features and appearance embedding features of the detected objects in consecutive frames. Effectively and robustly representing the spatial and appearance features of long trajectories has become a critical factor affecting the performance of MOT. We propose a novel approach for appearance and spatial-temporal motion feature representation, improving upon the hierarchical clustering association method MOT FCG. For spatialtemporal motion features, we first propose Diagonal Modulated GIoU, which more accurately represents the relationship between the position and shape of the objects. Second, Mean Constant Velocity Modeling is proposed to reduce the effect of observation noise on target motion state estimation. For appearance features, we utilize a dynamic appearance representation that incorporates confidence information, enabling the trajectory appearance features to be more robust and global. Based on the baseline model MOT FCG, we have realized further improvements in the performance of all. we achieved 63.1 HOTA, 76.9 MOTA and 78.2 IDF1 on the MOT17 test set, and also achieved competitive performance on the MOT20 and DanceTrack sets.

Figures

Figures reproduced from arXiv: 2411.10028 by the authors.

Figure 1
Figure 1. Comparison of Different Trackers on the MOT17-test Set and MOT20-test Set in Terms [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Illustration of MOT FCG++ 3.2 Clustering Spatial Motion Features Representation In the association between current frame detections and tracking trajectories, IoU is often calculated to represent the spatial motion information of objects. As shown in [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Comparison between IoU and Diagonal Modulated GIoU [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Limitations of Median Feature Representation [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Illustrations of MOT17, MOT20, and DanceTrack [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Comparison of Different Appearance Feature Representations. [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: Tracking performance of MOT FCG++ on the MOT17-train with different N values. Selection of DA. We conducted ablation experiments on the representation methods for trajectory appearance features. In this context, ”Median App.” refers to the median element feature repres…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 38 canonical work pages

  1. [1]

    Bot-sort: Robust associations multi-pedestrian tracking

    Nir Aharon, Roy Orfaig, and Ben-Zion Bobrovsky. Bot-sort: Robust associations multi-pedestrian tracking. arXiv preprint arXiv:2206.14651 , 2022

    arXiv 2022

  2. [2]

    Simple online and realtime tracking

    Alex Bewley, Zongyuan Ge, Lionel Ott, Fabio Ramos, and Ben Upcroft. Simple online and realtime tracking. In 2016 IEEE international conference on image processing (ICIP) , pages 3464–3468. IEEE, 2016

    work page 2016

  3. [3]

    Introduction to random signals and applied kalman filtering: with matlab exercises and solutions

    Robert Grover Brown and Patrick YC Hwang. Introduction to random signals and applied kalman filtering: with matlab exercises and solutions. Introduction to random signals and applied Kalman filtering: with MATLAB exercises and solutions , 1997

    work page 1997

  4. [4]

    Observation- centric sort: Rethinking sort for robust multi-object tracking

    Jinkun Cao, Jiangmiao Pang, Xinshuo Weng, Rawal Khirodkar, and Kris Kitani. Observation- centric sort: Rethinking sort for robust multi-object tracking. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 9686–9696, 2023

    work page 2023

  5. [5]

    Strongsort: Make deepsort great again

    Yunhao Du, Zhicheng Zhao, Yang Song, Yanyun Zhao, Fei Su, Tao Gong, and Hongying Meng. Strongsort: Make deepsort great again. IEEE Transactions on Multimedia , 25:8725–8737, 2023

    2023

  6. [6]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. arXiv, 2017

    work page 2017

  7. [7]

    Transtrack: Multiple-object tracking with transformer

    Peize Sun, Yi Jiang, Rufeng Zhang, Enze Xie, and Ping Luo. Transtrack: Multiple-object tracking with transformer. 2020

    work page 2020

  8. [8]

    Motr: End-to-end multiple-object tracking withtransformer

    Fangao Zeng, Bin Dong, Yuang Zhang, Tiancai Wang, Xiangyu Zhang, and Yichen Wei. Motr: End-to-end multiple-object tracking withtransformer. In European Conference on Computer Vi- sion, 2022

    work page 2022

Show all 41 references
  1. [9]

    Trackformer: Multi-object tracking with transformers

    Tim Meinhardt, Alexander Kirillov, Laura Leal-Taixe, and Christoph Feichtenhofer. Trackformer: Multi-object tracking with transformers. 2021

    2021

  2. [10]

    Transcenter: Transformers with dense queries for multiple-object tracking

    Yihong Xu, Yutong Ban, Guillaume Delorme, Chuang Gan, Daniela Rus, and Xavier Alameda- Pineda. Transcenter: Transformers with dense queries for multiple-object tracking. 2021

    2021

  3. [11]

    Transmot: Spatial- temporal graph transformer for multiple object tracking

    Peng Chu, Jiang Wang, Quanzeng You, Haibin Ling, and Zicheng Liu. Transmot: Spatial- temporal graph transformer for multiple object tracking. arXiv e-prints , 2021

    2021

  4. [12]

    Simple online and realtime tracking with a deep association metric

    Nicolai Wojke, Alex Bewley, and Dietrich Paulus. Simple online and realtime tracking with a deep association metric. IEEE, 2017

    2017

  5. [13]

    Tracking without bells and whistles

    Philipp Bergmann, Tim Meinhardt, and Laura Leal-Taixe. Tracking without bells and whistles. IEEE, 2019

    2019

  6. [14]

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

    Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster r-cnn: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis & Machine Intelligence, 39(6):1137–1149, 2017

    2017

  7. [15]

    Graph networks for multiple object tracking

    Jiahe Li, Xu Gao, and Tingting Jiang. Graph networks for multiple object tracking. IEEE, 2020

    2020

  8. [16]

    Learning a neural solver for multiple object tracking

    Guillem Bras´ o and Laura Leal-Taix´ e. Learning a neural solver for multiple object tracking. 2019

    2019

  9. [17]

    How to train your deep multi-object tracker, 2019

    Yihong Xu, Aljosa Osep, Yutong Ban, Radu Horaud, Laura Leal-Taixe, and Xavier Alameda- Pineda. How to train your deep multi-object tracker, 2019

    2019

  10. [18]

    Siammot: Siamese multi-object tracking

    Bing Shuai, Andrew Berneshawi, Xinyu Li, Davide Modolo, and Joseph Tighe. Siammot: Siamese multi-object tracking. 2021

    2021

  11. [19]

    Fairmot: On the fairness of detection and re-identification in multiple object tracking

    Yifu Zhang, Chunyu Wang, Xinggang Wang, Wenjun Zeng, and Wenyu Liu. Fairmot: On the fairness of detection and re-identification in multiple object tracking. International Journal of Computer Vision , (11), 2021. 12

    2021

  12. [20]

    Towards real-time multi-object tracking

    Zhongdao Wang, Liang Zheng, Yixuan Liu, Yali Li, and Shengjin Wang. Towards real-time multi-object tracking. Springer, Cham , 2020

    2020

  13. [21]

    Retinatrack: Online single stage joint detection and tracking

    Zhichao Lu, Vivek Rathod, Ronny Votel, and Jonathan Huang. Retinatrack: Online single stage joint detection and tracking. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020

    2020

  14. [22]

    Chained-tracker: Chain- ing paired attentive regression results for end-to-end joint multiple-object detection and tracking

    Jinlong Peng, Changan Wang, Fangbin Wan, Yang Wu, and Yanwei Fu. Chained-tracker: Chain- ing paired attentive regression results for end-to-end joint multiple-object detection and tracking. 2020

    2020

  15. [23]

    Observation- centric sort: Rethinking sort for robust multi-object tracking

    Jinkun Cao, Xinshuo Weng, Rawal Khirodkar, Jiangmiao Pang, and Kris Kitani. Observation- centric sort: Rethinking sort for robust multi-object tracking. 2022

    2022

  16. [24]

    Multiple object tracking from appearance by hierarchically clustering tracklets

    Andreu Girbau, Ferran Marqu´ es, and Shin’ichi Satoh. Multiple object tracking from appearance by hierarchically clustering tracklets. arXiv preprint arXiv:2210.03355 , 2022

    2022 arXiv

  17. [25]

    Hybrid-sort: Weak cues matter for online multi-object tracking

    Mingzhan Yang, Guangxin Han, Bin Yan, Wenhua Zhang, Jinqing Qi, Huchuan Lu, and Dong Wang. Hybrid-sort: Weak cues matter for online multi-object tracking. In Proceedings of the AAAI Conference on Artificial Intelligence , volume 38, pages 6504–6512, 2024

    2024

  18. [26]

    Author correction: Scipy 1.0: fundamental algorithms for scientific computing in python

    Pauli Virtanen, Ralf Gommers, Travis E Oliphant, Matt Haberland, and Paul Van Mulbregt. Author correction: Scipy 1.0: fundamental algorithms for scientific computing in python. Nature Methods, 17(Suppl. 1):1–12, 2020

    2020

  19. [27]

    Mot16: A bench- mark for multi-object tracking

    Anton Milan, Laura Leal-Taixe, Ian Reid, Stefan Roth, and Konrad Schindler. Mot16: A bench- mark for multi-object tracking. 2016

    2016

  20. [28]

    Mot20: A benchmark for multi object tracking in crowded scenes

    Patrick Dendorfer, Hamid Rezatofighi, Anton Milan, Javen Shi, Daniel Cremers, Ian Reid, Stefan Roth, Konrad Schindler, and Laura Leal-Taix´ e. Mot20: A benchmark for multi object tracking in crowded scenes. 2020

    2020

  21. [29]

    Dancetrack: Multi-object tracking in uniform appearance and diverse motion

    Peize Sun, Jinkun Cao, Yi Jiang, Zehuan Yuan, Song Bai, Kris Kitani, and Ping Luo. Dancetrack: Multi-object tracking in uniform appearance and diverse motion. IEEE

  22. [30]

    Evaluating multiple object tracking performance: The clear mot metrics

    Bernardin Keni and Stiefelhagen Rainer. Evaluating multiple object tracking performance: The clear mot metrics. EURASIP Journal on Image and Video Processing , 2008(1), 2008

    2008

  23. [31]

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

    Ergys Ristani, Francesco Solera, Roger S Zou, Rita Cucchiara, and Carlo Tomasi. Performance measures and a data set for multi-target, multi-camera tracking. Springer, Cham , 2016

    2016

  24. [32]

    Hota: A higher order metric for evaluating multi-object tracking

    et al Luiten, JonathonOsep. Hota: A higher order metric for evaluating multi-object tracking. International Journal of Computer Vision , 129(2), 2021

    2021

  25. [33]

    Tracking objects as points

    Philipp Krhenb¨ uhl, Vladlen Koltun, and Xingyi Zhou. Tracking objects as points. In European Conference on Computer Vision , 2020

    2020

  26. [34]

    Yolox: Exceeding yolo series in

    Zheng Ge, Songtao Liu, Feng Wang, Zeming Li, and Jian Sun. Yolox: Exceeding yolo series in

  27. [35]

    Scalable person re-identification: A benchmark

    Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, and Qi Tian. Scalable person re-identification: A benchmark. In 2015 IEEE International Conference on Computer Vision (ICCV) , 2015

    2015

  28. [36]

    Fastreid: A py- torch toolbox for general instance re-identification

    Lingxiao He, Xingyu Liao, Wu Liu, Xinchen Liu, Peng Cheng, and Tao Mei. Fastreid: A py- torch toolbox for general instance re-identification. In Proceedings of the 31st ACM International Conference on Multimedia, MM ’23, page 9664–9667. Association for Computing Machinery, 2023

    2023

  29. [37]

    Unweighted pair group method with arithmetic mean (upgma)

    DB Hibbert. Unweighted pair group method with arithmetic mean (upgma). IUPAC Stand Online, 10, 2017

    2017

  30. [38]

    Real-time multiple people tracking with deeply learned candidate selection and person re-identification

    Long Chen, Haizhou Ai, Zijie Zhuang, and Chong Shang. Real-time multiple people tracking with deeply learned candidate selection and person re-identification. In IEEE International Conference on Multimedia & Expo , pages 1–6, 2018. 13

    2018

  31. [39]

    Ucmctrack: Multi-object tracking with uniform camera motion compensation

    Kefu Yi, Kai Luo, Xiaolei Luo, Jiangui Huang, Hao Wu, Rongdong Hu, and Wei Hao. Ucmctrack: Multi-object tracking with uniform camera motion compensation. In Proceedings of the AAAI Conference on Artificial Intelligence , volume 38, pages 6702–6710, 2024

    2024

  32. [40]

    Bytetrack: Multi-object tracking by associating every detection box

    Y Zhang, P Sun, Y Jiang, D Yu, F Weng, Z Yuan, P Luo, W Liu, and X Wang. Bytetrack: Multi-object tracking by associating every detection box. 2022. 14

    2022

  33. [2021]

    arXiv e-prints , 2021

    2021

Pith tools

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