pith. sign in

arxiv: 2606.30476 · v1 · pith:XJBKZLNCnew · submitted 2026-06-29 · 💻 cs.CV · cs.RO· eess.IV

PS-MOT: Cultivating Instance Awareness from Point Seeds for Multi-Object Tracking

Kai Luo , Fei Teng , Mengfei Duan , Wanjun Jia , Xu Wang , Hao Shi , Kunyu Peng , Zhiyong Li
show 1 more author
Kailun Yang
This is my paper

Pith reviewed 2026-06-30 06:43 UTC · model grok-4.3

classification 💻 cs.CV cs.ROeess.IV
keywords point-supervised multi-object trackingpseudo-label generationtemporal promptingwavelet attentionuncertainty-guided learninginstance awareness
0
0 comments X

The pith

Point seeds can be evolved into instance-aware multi-object trackers via temporal prompting, wavelet attention, and uncertainty-guided loss.

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

The paper sets out to prove that multi-object tracking can be done effectively with only point annotations rather than full bounding boxes. It introduces a pipeline that turns sparse point seeds into consistent pseudo-labels, activates object boundaries in the model, and treats labels as probabilistic distributions during training. If the approach holds, tracking systems could operate with far less annotation effort while maintaining competitive accuracy across varied video domains.

Core claim

PS-Track forms a hierarchical pipeline that cultivates instance awareness from point seeds at data, model, and loss levels. Temporal-Feedback Prompting generates temporally consistent pseudo-labels from points using motion priors and negative cues. Point-Excited Wavelet Attention activates high-frequency components to hallucinate boundaries from semantic correlations. Uncertainty-Guided Gaussian Learning models the pseudo-labels as distributions that dynamically adjust supervision strength, yielding state-of-the-art results for point-supervised tracking on DanceTrack, EmboTrack, SportsMOT, and JRDB.

What carries the argument

The PS-Track hierarchical pipeline that transitions from points to instances using Temporal-Feedback Prompting at the data level, Point-Excited Wavelet Attention at the model level, and Uncertainty-Guided Gaussian Learning at the loss level.

If this is right

  • Point supervision becomes a practical substitute for bounding-box supervision in multi-object tracking.
  • The same pipeline delivers leading results on dance, sports, and pedestrian-robot datasets under point-only labels.
  • Annotation effort for training trackers can shift from drawing boxes to marking centers without loss of capability.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same point-to-instance cultivation steps could be tested on single-object tracking or video instance segmentation to check transfer.
  • If the pseudo-label evolution proves stable, the method may reduce the data cost of building trackers for new environments by an order of magnitude.
  • Extending the uncertainty model to handle long-term occlusions would be a direct next measurement on the same benchmarks.

Load-bearing premise

The three components together can resolve spatial ambiguity and identity drift when the only input is point seeds that carry no explicit geometric or scale information.

What would settle it

On a new video dataset or ablation setting, removing any one of the three components causes the tracker to fall below prior point-supervised methods or to exhibit clear rises in identity switches and localization errors.

Figures

Figures reproduced from arXiv: 2606.30476 by Fei Teng, Hao Shi, Kailun Yang, Kai Luo, Kunyu Peng, Mengfei Duan, Wanjun Jia, Xu Wang, Zhiyong Li.

Figure 1
Figure 1. Figure 1: Comparison between Fully Supervised MOT and our proposed Point [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Versatility of PS￾Track across mainstream tracking paradigms. labels, we transition from deterministic regression to Uncertainty-Guided Gaussian Learning (UGL) at the loss level. Since point annotations lack physical ex￾tents, deterministic box regression causes noise mem￾orization. UGL circumvents this by modeling pseudo￾labels as Gaussian distributions, naturally accommo￾dating spatial ambiguity. To miti… view at source ↗
Figure 3
Figure 3. Figure 3: The overview of our proposed framework. It operates on a coarse-to-fine paradigm across three levels: (a) Data Level: The Temporal-Feedback Prompting mechanism evolves sparse points into consistent pseudo-labels; (b) Model Level: The Point-Excited Wavelet Attention (PEWA) module leverages frequency decomposition to Hallucinate boundaries from point cues; (c) Loss Level: The Uncertainty-Guided Gaussian Loss… view at source ↗
Figure 4
Figure 4. Figure 4: Visual comparison of pseudo-labels generated by the original SAM [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative comparison on the DanceTrack dataset. [PITH_FULL_IMAGE:figures/full_fig_p015_5.png] view at source ↗
read the original abstract

We introduce Point-supervised Multi-Object Tracking (PS-MOT) as a cost-effective alternative to traditional bounding box supervision, shifting the focus from spatial fitting to topological center-driven representation. However, PS-MOT faces challenges, e.g., spatial ambiguity and identity drift due to the lack of explicit geometric structure and scale constraints. To address these, we propose PS-Track, a hierarchical pipeline transitioning from points to instances across data, model, and loss levels. At the data level, we introduce Temporal-Feedback Prompting (TFP) to evolve points into temporally consistent pseudo-labels using negative spatial cues and motion priors. At the model level, we design the Point-Excited Wavelet Attention (PEWA) module, which leverages semantic correlations to activate high-frequency components, ``hallucinating'' object boundaries. At the loss level, Uncertainty-Guided Gaussian Learning (UGL) models pseudo-labels as probabilistic distributions, dynamically calibrating supervision intensity. Experiments on DanceTrack, EmboTrack, SportsMOT, and JRDB demonstrate that PS-Track provides a feasible and effective point-supervised alternative across diverse tracking scenarios, establishing a new state-of-the-art for point-supervised tracking. The source code is available at https://github.com/xifen523/PS-MOT.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 2 minor

Summary. The paper introduces Point-supervised Multi-Object Tracking (PS-MOT) as a cost-effective alternative to bounding box supervision, focusing on topological center-driven representation from point seeds. It proposes the PS-Track hierarchical pipeline with three components: Temporal-Feedback Prompting (TFP) at the data level to evolve points into temporally consistent pseudo-labels via negative spatial cues and motion priors; Point-Excited Wavelet Attention (PEWA) at the model level to leverage semantic correlations for activating high-frequency components and hallucinating object boundaries; and Uncertainty-Guided Gaussian Learning (UGL) at the loss level to model pseudo-labels as probabilistic distributions for dynamic supervision calibration. Experiments across DanceTrack, EmboTrack, SportsMOT, and JRDB are reported to demonstrate feasibility and establish a new state-of-the-art for point-supervised tracking, with source code released.

Significance. If the empirical claims hold, the work provides a practical reduction in annotation cost for MOT by shifting to point supervision while addressing the specific challenges of spatial ambiguity and identity drift through a structured data-model-loss pipeline. The open release of code supports reproducibility. This could meaningfully expand research directions in weakly-supervised tracking by showing that point seeds can suffice when augmented with the proposed mechanisms for pseudo-label evolution and boundary inference.

minor comments (2)
  1. [Abstract] Abstract: the SOTA claim is stated without any quantitative metrics, baseline comparisons, or dataset-specific scores; adding one or two key numbers (e.g., HOTA or MOTA deltas) would make the summary self-contained.
  2. [Abstract] Abstract and §3 (model level): the phrase 'hallucinating object boundaries' is used without a precise technical definition or reference to how PEWA's wavelet activation produces explicit boundary outputs versus implicit feature enhancement; a short clarification or diagram reference would improve precision.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the thorough summary of our work, the positive assessment of its significance, and the recommendation for minor revision. No specific major comments were provided in the report.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper introduces PS-Track as a hierarchical pipeline with three explicitly proposed modules (Temporal-Feedback Prompting at data level, Point-Excited Wavelet Attention at model level, and Uncertainty-Guided Gaussian Learning at loss level) to address stated challenges of point supervision. These components are presented as novel designs whose effectiveness is validated through experiments on four external datasets, with source code released. No equations, self-citations, or claims reduce any prediction or result to a fitted parameter or definition internal to the same work by construction; the derivation chain remains self-contained and externally falsifiable via the reported benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract supplies no equations, training details, or modeling choices, so no free parameters, axioms, or invented entities can be identified.

pith-pipeline@v0.9.1-grok · 5787 in / 1097 out tokens · 31291 ms · 2026-06-30T06:43:53.338041+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

72 extracted references · 12 canonical work pages · 3 internal anchors

  1. [1]

    BoT-SORT: Robust associations multi-pedestrian tracking,

    Aharon, N., Orfaig, R., Bobrovsky, B.Z.: BoT-SORT: Robust associations multi- pedestrian tracking. arXiv preprint arXiv:2206.14651 (2022)

    work page arXiv 2022

  2. [2]

    Journal of Cognitive Neuroscience (2003)

    Bar, M.: A cortical mechanism for triggering top-down facilitation in visual object recognition. Journal of Cognitive Neuroscience (2003)

    2003

  3. [3]

    In: ECCV (2016)

    Bearman, A., Russakovsky, O., Ferrari, V., Fei-Fei, L.: What’s the point: Semantic segmentation with point supervision. In: ECCV (2016)

    2016

  4. [4]

    EURASIP Journal on Image and Video Processing (2008)

    Bernardin, K., Stiefelhagen, R.: Evaluating multiple object tracking performance: The CLEAR MOT metrics. EURASIP Journal on Image and Video Processing (2008)

    2008

  5. [5]

    In: ICIP (2016)

    Bewley, A., Ge, Z., Ott, L., Ramos, F., Upcroft, B.: Simple online and realtime tracking. In: ICIP (2016)

    2016

  6. [6]

    In: CVPR (2016)

    Bilen, H., Vedaldi, A.: Weakly supervised deep detection networks. In: CVPR (2016)

    2016

  7. [7]

    In: CVPR (2023)

    Cao, J., Pang, J., Weng, X., Khirodkar, R., Kitani, K.: Observation-centric SORT: Rethinking SORT for robust multi-object tracking. In: CVPR (2023)

    2023

  8. [8]

    SAM 3: Segment Anything with Concepts

    Carion, N., Gustafson, L., Hu, Y.T., Debnath, S., Hu, R., Suris, D., Ryali, C., Alwala, K.V., Khedr, H., Huang, A., Lei, J., Ma, T., Guo, B., Kalla, A., Marks, M., Greer, J., Wang, M., Sun, P., Rädle, R., Afouras, T., Mavroudi, E., Xu, K., Wu, T.H., Zhou, Y., Momeni, L., Hazra, R., Ding, S., Vaze, S., Porcher, F., Li, F., Li, S., Kamath, A., Cheng, H.K., ...

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  9. [9]

    In: CVPR (2023)

    Cetintas, O., Brasó, G., Leal-Taixé, L.: Unifying short and long-term tracking with graph hierarchies. In: CVPR (2023)

    2023

  10. [10]

    Scientific Reports (2025)

    Chan, S., Zhou, W., Lei, Y., Li, C., Hu, J., Hong, F.: Sparse point annotations for remote sensing image segmentation. Scientific Reports (2025)

    2025

  11. [11]

    In: ECCV (2022)

    Chen, P., Yu, X., Han, X., Hassan, N., Wang, K., Li, J., Zhao, J., Shi, H., Han, Z., Ye, Q.: Point-to-box network for accurate object detection via single point supervision. In: ECCV (2022)

    2022

  12. [12]

    In: CVPR (2022)

    Cheng, B., Parkhi, O., Kirillov, A.: Pointly-supervised instance segmentation. In: CVPR (2022)

    2022

  13. [13]

    In: ICCV (2023)

    Cui, Y., Zeng, C., Zhao, X., Yang, Y., Wu, G., Wang, L.: SportsMOT: A large multi-object tracking dataset in multiple sports scenes. In: ICCV (2023)

    2023

  14. [14]

    International Journal of Computer Vision (2021)

    Dendorfer, P., Osep, A., Milan, A., Schindler, K., Cremers, D., Reid, I., Roth, S., Leal-Taixé, L.: MOTChallenge: A benchmark for single-camera multiple target tracking. International Journal of Computer Vision (2021)

    2021

  15. [15]

    arXiv preprint arXiv:2003.09003 (2020)

    Dendorfer, P., Rezatofighi, H., Milan, A., Shi, J., Cremers, D., Reid, I., Roth, S., Schindler, K., Leal-Taixé, L.: MOT20: A benchmark for multi object tracking in crowded scenes. arXiv preprint arXiv:2003.09003 (2020)

    work page arXiv 2003

  16. [16]

    In: ICML (2016)

    Gal, Y., Ghahramani, Z.: Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. In: ICML (2016)

    2016

  17. [17]

    In: CVPR (2025)

    Gao, R., Qi, J., Wang, L.: Multiple object tracking as ID prediction. In: CVPR (2025)

    2025

  18. [18]

    In: ICCV (2023)

    Gao, R., Wang, L.: MeMOTR: Long-term memory-augmented transformer for multi-object tracking. In: ICCV (2023)

    2023

  19. [19]

    IEEE Transactions on Circuits and Systems for Video Technology (2025) PS-MOT 17

    Gao, X., Li, Z., Shi, H., Chen, Z., Zhao, P.: Scribble-supervised video object seg- mentation via scribble enhancement. IEEE Transactions on Circuits and Systems for Video Technology (2025) PS-MOT 17

    2025

  20. [20]

    YOLOX: Exceeding YOLO Series in 2021

    Ge, Z., Liu, S., Wang, F., Li, Z., Sun, J.: YOLOX: Exceeding YOLO series in 2021. arXiv preprint arXiv:2107.08430 (2021)

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  21. [21]

    IEEE Access (2025)

    Hayat, M., Aramvith, S.: Superpixel-guided graph-attention boundary GAN for adaptive feature refinement in scribble-supervised medical image segmentation. IEEE Access (2025)

    2025

  22. [22]

    In: CVPR (2021)

    He, J., Huang, Z., Wang, N., Zhang, Z.: Learnable graph matching: Incorporat- ing graph partitioning with deep feature learning for multiple object tracking. In: CVPR (2021)

    2021

  23. [23]

    In: ICCV (2017)

    He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask R-CNN. In: ICCV (2017)

    2017

  24. [24]

    In: CVPR (2016)

    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: CVPR (2016)

    2016

  25. [25]

    In: CVPR (2023)

    Hu, Y., Yang, J., Chen, L., Li, K., Sima, C., Zhu, X., Chai, S., Du, S., Lin, T., Wang, W., Lu, L., Jia, X., Liu, Q., Dai, J., Qiao, Y., Li, H.: Planning-oriented autonomous driving. In: CVPR (2023)

    2023

  26. [26]

    arXiv preprint arXiv:2601.01925 (2026)

    Jia, L., Wu, Y., Ran, B., Wang, Y., Wang, L., Lu, H.: AR-MOT: Autoregressive multi-object tracking. arXiv preprint arXiv:2601.01925 (2026)

    work page arXiv 2026

  27. [27]

    Kendall, A., Gal, Y.: What uncertainties do we need in Bayesian deep learning for computer vision? In: NeurIPS (2017)

    2017

  28. [28]

    Knowledge and Information Systems (2025)

    Li, S., Yang, L., Tan, H., Wang, B., Huang, W., Liu, H., Yang, W., Lan, L.: Self- supervised re-identification for online joint multi-object tracking. Knowledge and Information Systems (2025)

    2025

  29. [29]

    arXiv preprint arXiv:2411.06702 (2024)

    Lim, J.S., Luo, Y., Chen, Z., Wei, T., Chapman, S., Huang, Z.: Track any peppers: Weakly supervised sweet pepper tracking using VLMs. arXiv preprint arXiv:2411.06702 (2024)

    work page arXiv 2024

  30. [30]

    In: CVPR (2024)

    Lu, Z., Shuai, B., Chen, Y., Xu, Z., Modolo, D.: Self-supervised multi-object track- ing with path consistency. In: CVPR (2024)

    2024

  31. [31]

    Interna- tional Journal of Computer Vision (2021)

    Luiten, J., Osep, A., Dendorfer, P., Torr, P.H.S., Geiger, A., Leal-Taixé, L., Leibe, B.: HOTA: A higher order metric for evaluating multi-object tracking. Interna- tional Journal of Computer Vision (2021)

    2021

  32. [32]

    OmniTrack++: Omnidirectional Multi-Object Tracking by Learning Large-FoV Trajectory Feedback

    Luo, K., Shi, H., Peng, K., Teng, F., Wu, S., Wang, K., Yang, K.: OmniTrack++: Omnidirectional multi-object tracking by learning large-FoV trajectory feedback. arXiv preprint arXiv:2511.00510 (2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  33. [33]

    In: CVPR (2025)

    Luo, K., Shi, H., Wu, S., Teng, F., Duan, M., Huang, C., Wang, Y., Wang, K., Yang, K.: Omnidirectional multi-object tracking. In: CVPR (2025)

    2025

  34. [34]

    In: CVPR (2024)

    Lv, W., Huang, Y., Zhang, N., Lin, R.S., Han, M., Zeng, D.: DiffMOT: A real- time diffusion-based multiple object tracker with non-linear prediction. In: CVPR (2024)

    2024

  35. [35]

    In: NeurIPS (2021)

    Mao, J., Niu, M., Jiang, C., Liang, H., Chen, J., Liang, X., Li, Y., Ye, C., Zhang, W., Li, Z., Yu, J., Xu, H., Xu, C.X.: One million scenes for autonomous driving: Once dataset. In: NeurIPS (2021)

    2021

  36. [36]

    IEEE Transactions on Pattern Analysis and Machine Intelligence (2023)

    Martin-Martin, R., Patel, M., Rezatofighi, H., Shenoi, A., Gwak, J., Frankel, E., Sadeghian, A., Savarese, S.: JRDB: A dataset and benchmark of egocentric robot visual perception of humans in built environments. IEEE Transactions on Pattern Analysis and Machine Intelligence (2023)

    2023

  37. [37]

    In: CVPR (2022)

    Meinhardt, T., Kirillov, A., Leal-Taixé, L., Feichtenhofer, C.: TrackFormer: Multi- object tracking with transformers. In: CVPR (2022)

    2022

  38. [38]

    In: CVPR (2021)

    Pang,J., Qiu,L., Li,X., Chen,H., Li,Q., Darrell,T., Yu, F.:Quasi-dense similarity learning for multiple object tracking. In: CVPR (2021)

    2021

  39. [39]

    In: ICCV (2017) 18 K

    Papadopoulos, D.P., Uijlings, J.R.R., Keller, F., Ferrari, V.: Extreme clicking for efficient object annotation. In: ICCV (2017) 18 K. Luoet al

    2017

  40. [40]

    arXiv preprint arXiv:2410.13842 (2024)

    Peng, Y., Li, H., Wu, P., Zhang, Y., Sun, X., Wu, F.: D-FINE: Redefine re- gression task in detrs as fine-grained distribution refinement. arXiv preprint arXiv:2410.13842 (2024)

    work page arXiv 2024

  41. [41]

    In: CVPR (2023)

    Ren, H., Han, S., Ding, H., Zhang, Z., Wang, H., Wang, F.: Focus on details: Online multi-object tracking with diverse fine-grained representation. In: CVPR (2023)

    2023

  42. [42]

    In: ECCVW (2016)

    Ristani, E., Solera, F., Zou, R., Cucchiara, R., Tomasi, C.: Performance measures and a data set for multi-target, multi-camera tracking. In: ECCVW (2016)

    2016

  43. [43]

    IEEE Transactions on Signal Processing (2002)

    Shensa, M.J.: The discrete wavelet transform: wedding the a trous and Mallat algorithms. IEEE Transactions on Signal Processing (2002)

    2002

  44. [44]

    In: ICIP (2024)

    Shim, K., Hwang, J., Ko, K., Kim, C.: A confidence-aware matching strategy for generalized multi-object tracking. In: ICIP (2024)

    2024

  45. [45]

    In: CVPR (2025)

    Shim, K., Ko, K., Yang, Y., Kim, C.: Focusing on tracks for online multi-object tracking. In: CVPR (2025)

    2025

  46. [46]

    In: AAAI (2026)

    Shu, Z., Wu, J., Yan, W., Liu, X., Zhang, H., Liu, C., Mao, Y., Chen, J.: Wave- Former: Frequency-time decoupled vision modeling with wave equation. In: AAAI (2026)

    2026

  47. [47]

    In: CVPR (2022)

    Sun, P., Cao, J., Jiang, Y., Yuan, Z., Bai, S., Kitani, K., Luo, P.: DanceTrack: Multi-object tracking in uniform appearance and diverse motion. In: CVPR (2022)

    2022

  48. [48]

    arXiv preprint arXiv:2012.15460 (2020)

    Sun, P., Cao, J., Jiang, Y., Zhang, R., Xie, E., Yuan, Z., Wang, C., Luo, P.: TransTrack: Multiple object tracking with transformer. arXiv preprint arXiv:2012.15460 (2020)

    work page arXiv 2012

  49. [49]

    In: AAAI (2024)

    Tan, C., Zhao, Y., Wei, S., Gu, G., Liu, P., Wei, Y.: Frequency-aware deepfake detection: Improving generalizability through frequency space domain learning. In: AAAI (2024)

    2024

  50. [50]

    In: CVPR (2017)

    Tang, P., Wang, X., Bai, X., Liu, W.: Multiple instance detection network with online instance classifier refinement. In: CVPR (2017)

    2017

  51. [51]

    Cerebral Cortex (1995)

    Ullman, S.: Sequence seeking and counter streams: a computational model for bidirectional information flow in the visual cortex. Cerebral Cortex (1995)

    1995

  52. [52]

    arXiv preprint arXiv:2411.08433 (2024)

    Wang, X., Liu, J., Feng, M., Zhang, Z., Yang, X.: 3D multi-object tracking with semi-supervised GRU-Kalman filter. arXiv preprint arXiv:2411.08433 (2024)

    work page arXiv 2024

  53. [53]

    In: ICIP (2017)

    Wojke, N., Bewley, A., Paulus, D.: Simple online and realtime tracking with a deep association metric. In: ICIP (2017)

    2017

  54. [54]

    In: CVPR (2021)

    Wu, J., Cao, J., Song, L., Wang, Y., Yang, M., Yuan, J.: Track to detect and segment: An online multi-object tracker. In: CVPR (2021)

    2021

  55. [55]

    In: AAAI (2024)

    Yang, M., Han, G., Yan, B., Zhang, W., Qi, J., Lu, H., Wang, D.: Hybrid-SORT: Weak cues matter for online multi-object tracking. In: AAAI (2024)

    2024

  56. [56]

    In: AAAI (2024)

    Yi, K., Luo, K., Luo, X., Huang, J., Wu, H., Hu, R., Hao, W.: UCMCTrack: Multi- object tracking with uniform camera motion compensation. In: AAAI (2024)

    2024

  57. [57]

    In: CVPR (2022)

    Yu, X., Chen, P., Wu, D., Hassan, N., Li, G., Yan, J., Shi, H., Ye, Q., Han, Z.: Object localization under single coarse point supervision. In: CVPR (2022)

    2022

  58. [58]

    In: CVPR (2025)

    Yu, Y., Ren, B., Zhang, P., Liu, M., Luo, J., Zhang, S., Da, F., Yan, J., Yang, X.: Point2RBox-v2: Rethinking point-supervised oriented object detection with spatial layout among instances. In: CVPR (2025)

    2025

  59. [59]

    In: CVPR (2024)

    Yu, Y., Yang, X., Li, Q., Da, F., Dai, J., Qiao, Y., Yan, J.: Point2RBox: Combine knowledge from synthetic visual patterns for end-to-end oriented object detection with single point supervision. In: CVPR (2024)

    2024

  60. [60]

    In: ECCV (2022) PS-MOT 19

    Zeng, F., Dong, B., Zhang, Y., Wang, T., Zhang, X., Wei, Y.: MOTR: End-to-end multiple-object tracking with transformer. In: ECCV (2022) PS-MOT 19

    2022

  61. [61]

    arXiv preprint arXiv:2509.26281 (2025)

    Zhang,T.,Fan,Z.,Liu,M.,Zhang,X.,Lu,X.,Li,W.,Zhou,Y.,Yu,Y.,Li,X.,Yan, J., Yang, X.: Point2RBox-v3: Self-bootstrapping from point annotations via inte- grated pseudo-label refinement and utilization. arXiv preprint arXiv:2509.26281 (2025)

    work page arXiv 2025

  62. [62]

    In: ECCV (2022)

    Zhang, Y., Sun, P., Jiang, Y., Yu, D., Weng, F., Yuan, Z., Luo, P., Liu, W., Wang, X.: ByteTrack: Multi-object tracking by associating every detection box. In: ECCV (2022)

    2022

  63. [63]

    International Journal of Computer Vision (2021)

    Zhang, Y., Wang, C., Wang, X., Zeng, W., Liu, W.: FairMOT: On the fairness of detection and re-identification in multiple object tracking. International Journal of Computer Vision (2021)

    2021

  64. [64]

    In: CVPR (2023)

    Zhang, Y., Wang, T., Zhang, X.: MOTRv2: Bootstrapping end-to-end multi-object tracking by pretrained object detectors. In: CVPR (2023)

    2023

  65. [65]

    arXiv preprint arXiv:2502.16809 (2025)

    Zhao, Z., Yu, J., Zhang, L., Zhang, S.: CRTrack: Low-light semi-supervised multi-object tracking based on consistency regularization. arXiv preprint arXiv:2502.16809 (2025)

    work page arXiv 2025

  66. [66]

    In: AAAI (2025)

    Zheng, M., Xu, Z., Xia, Q., Wu, H., Wen, C., Wang, C.: Seg2Box: 3D object detection by point-wise semantics supervision. In: AAAI (2025)

    2025

  67. [67]

    IEEE Transactions on Geoscience and Remote Sensing (2022)

    Zheng, S., Wu, Z., Xu, Y., Wei, Z., Plaza, A.: Learning orientation information from frequency-domain for oriented object detection in remote sensing images. IEEE Transactions on Geoscience and Remote Sensing (2022)

    2022

  68. [68]

    In: ECCV (2020)

    Zhou, X., Koltun, V., Krähenbühl, P.: Tracking objects as points. In: ECCV (2020)

    2020

  69. [69]

    In: CVPR (2022)

    Zhou, X., Yin, T., Koltun, V., Krähenbühl, P.: Global tracking transformers. In: CVPR (2022)

    2022

  70. [70]

    IEEE Transactions on Intelligent Vehicles (2024)

    Zhou, Y., Cai, L., Cheng, X., Zhang, Q., Xue, X., Ding, W., Pu, J.: OpenAnno- tate2: Multi-modal auto-annotating for autonomous driving. IEEE Transactions on Intelligent Vehicles (2024)

    2024

  71. [71]

    IEEE Transactions on Artificial Intelligence (2025)

    Zhu, R., Zhao, J., Zhang, D., Wang, G., Chen, X., Zhang, S., Gong, J., Zhou, Q., Zhang, W., Wang, N., Tan, F., Xu, Z., Zhou, H., Yao, H., Zhang, C., Liu, L., Liu, X., Di, X., Li, B.: SparseAD: Sparse query-centric paradigm for efficient end-to-end autonomous driving. IEEE Transactions on Artificial Intelligence (2025)

    2025

  72. [72]

    In: ICLR (2021) 20 K

    Zhu, X., Su, W., Lu, L., Li, B., Wang, X., Dai, J.: Deformable DETR: Deformable transformers for end-to-end object detection. In: ICLR (2021) 20 K. Luoet al. A Implementation Details Detailed Network Architecture.We implement PS-Track built upon the MOTIP [17] framework, adopting Deformable DETR [72] as our core detec- tor. The visual features are extract...

    2021