Pith. sign in

REVIEW 5 major objections 5 minor 2 cited by

PoLaRIS Dataset: A Maritime Object Detection and Tracking Dataset in Pohang Canal

T0 review · 5 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read New dataset tags ships and buoys in RGB, thermal, LiDAR, and radar.

desk verdict A genuinely new multi-modal maritime dataset, but the TIR label-generation formula is geometrically unsound as stated and label quality is unquantified; worth reviewing with conditions. read the letter →

arxiv 2412.06192 v2 pith:A5GUBBN4 submitted 2024-12-09 cs.RO

Jiwon Choi , Dongjin Cho , Gihyeon Lee , Hogyun Kim , Geonmo Yang , Joowan Kim , Younggun Cho This is my paper · ORCID
classification cs.RO
keywords maritimeobjectdetectionmulti-modaldatasetsmalltrackingthermalinfraredLiDARpointcloudradarunmannedsurfacevehicle
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 introduces PoLaRIS, a maritime obstacle dataset recorded in Pohang Canal, and claims it is the first to pair RGB and thermal camera bounding boxes with point-wise LiDAR and radar annotations and tracking IDs for dynamic obstacles. The authors argue that existing maritime datasets lack depth information and tracking labels, especially for small distant objects, which limits work on collision avoidance for unmanned surface vehicles. The dataset provides roughly 190,000 labels across five day and night sequences, including objects as small as 10 by 10 pixels. Benchmark runs with standard detectors and trackers on RGB and TIR imagery show strong daytime RGB performance and reveal that open-vocabulary detectors struggle with the small objects.

What carries the argument

The load-bearing mechanism is the semi-automatic multi-modal annotation pipeline. Left RGB bounding boxes are produced by detector initialization plus manual cleanup, then propagated to the TIR camera through the projective transformation in Equation 2 and to LiDAR by filtering for points whose projections fall inside the RGB box. Radar labels come from converting labeled LiDAR points into radar coordinates, projecting both into bird's-eye view, and applying DBSCAN clustering so radar clusters overlapping the LiDAR labels inherit the label. This pipeline turns one manual 2D annotation into synchronized thermal, LiDAR, and radar labels across sensors.

What would settle it

Take a scene with a calibration target visible to all four sensors, project its corners from the left camera into the TIR image and LiDAR/radar coordinates using the published pipeline, and measure the offset against manually labeled ground truth; if the offset is comparable to or larger than the smallest annotated objects (10 pixels) or causes a substantial fraction of LiDAR/radar points to fall on non-object surfaces, the cross-modal labels at the claimed scale are not reliable.

Watch

Extended reading notes

Core claim

The central claim is that PoLaRIS is the first maritime object detection and tracking dataset to provide simultaneous annotations in RGB, thermal infrared, LiDAR, and radar, including per-point depth labels and tracking ground truth for ships and buoys. Labeling begins on left RGB frames: initial boxes from a detector are manually refined, then projected into the TIR image with the extrinsic calibration transformation $bb_t = K_t T^t_l K_l^{-1} bb_l$, while LiDAR points are kept only when they project inside the RGB bounding box. Radar points are clustered with DBSCAN and labeled by overlap with the LiDAR annotations in bird's-eye view. Evaluations with YOLOv8, YOLOv10, RT-DETR, SORT, ByteTrack, OC-SORT, and Hybrid-SORT on day and night, RGB and TIR data are presented as evidence that the dataset supports detection and tracking benchmarking.

Load-bearing premise

The cross-modal labels are only as correct as the extrinsic calibration and temporal synchronization among the RGB, TIR, LiDAR, and radar sensors, and the paper does not quantify how calibration error propagates into the projected boxes, LiDAR filtering, or radar clusters.

Editorial extensions

If this is right

  • Detectors and trackers can now be evaluated on identical maritime scenes across visible, thermal, LiDAR, and radar modalities, enabling direct cross-sensor fusion studies.
  • The 10-by-10-pixel annotations provide a small-object benchmark; in the reported tests, open-vocabulary detectors score below 2.5 mAP without fine-tuning, showing where generalization currently fails.
  • Tracking IDs for dynamic objects make it possible to measure identity preservation (MOTA, IDF1) on small maritime obstacles rather than only detection accuracy.
  • Night-sequence labels produced with low-light restoration and TIR projection allow quantitative comparison of RGB and thermal performance under poor visibility.

Reading between the lines

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

  • Because radar labels are derived only from clusters overlapping LiDAR points inside RGB boxes, the dataset likely under-labels distant radar returns beyond LiDAR range; users testing radar-only long-range detection should treat the radar annotations as bounded by LiDAR coverage.
  • If the extrinsic calibration between cameras drifts, the projected TIR boxes and LiDAR filtering inherit that error; a simple check would be comparing projected TIR boxes against the manually corrected versions the authors describe, since the paper does not report this agreement.
  • The same semi-automatic pipeline could be reused to add multi-modal labels to other single-modality maritime datasets, provided the sensors can be calibrated and synchronized.
  • The paper reports overall tracking scores but does not break them down by object size; analyzing MOTA and identity switches for only the smallest objects would test whether the dataset actually supports the claimed small-object tracking contribution.
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

5 major / 5 minor

Summary. The paper introduces PoLaRIS, a maritime object detection and tracking dataset built on the Pohang Canal dataset. It provides left/right RGB, TIR, LiDAR, and Radar annotations for five sequences, including tracking IDs for dynamic objects and annotations for small objects down to 10×10 pixels. A semi-automatic pipeline initializes RGB labels with YOLOv8 and manual refinement, transfers labels to TIR via extrinsics, filters LiDAR points projected into boxes, and assigns Radar clusters by overlap with LiDAR labels. Benchmarks with YOLOv8/YOLOv10/RT-DETR and SORT/ByteTrack/OC-SORT/Hybrid-SORT are reported for RGB and TIR. The paper claims to be the first maritime dataset with multi-modal annotations across these four modalities.

Significance. If the annotation quality is established, PoLaRIS fills a genuine gap: no existing maritime dataset provides synchronized RGB, TIR, LiDAR, and Radar labels with tracking IDs and small-object coverage. The resource is potentially valuable for USV perception, and the benchmark numbers for RGB/TIR detection and tracking are useful. Strengths include the public dataset, use of an existing raw-data collection (Pohang Canal), a semi-automatic pipeline with human verification, and evaluation with several SOTA detectors and trackers. The construction is independent of the benchmark results, so there is no circularity in the evaluation. However, the paper currently lacks quantitative validation of the transferred and point-wise labels, so the significance depends on additional evidence.

major comments (5)
  1. [§III.B.1, Eq. (2)] Equation (2) is not a valid perspective transfer unless the TIR-left baseline is zero. Since bb_l is the homogeneous pixel coordinate [u v 1]^T, K_l^{-1} bb_l is the ray direction at depth 1, and applying T_t^l (which contains translation) projects that point as though every object lies at depth 1 m. The correct transfer needs per-object depth Z: bb_t = K_t (R (Z K_l^{-1} bb_l) + t). The manual post-processing mentioned in the text may compensate, but no quantitative measure of TIR label quality (e.g., alignment error, correction rate, or inter-annotator agreement) is reported. Because TIR annotations are a core contribution, this must be addressed.
  2. [§III.B.3] The Radar annotation procedure relies on DBSCAN clustering and assigns labels to clusters that overlap with LiDAR labels, but the paper does not report the DBSCAN parameters (eps, min_samples), the overlap criterion, or any validation of the resulting point labels. Given the wide horizontal beamwidth of marine Radar, cluster-level overlap can merge distinct objects or include clutter. The paper should report parameters and a quantitative assessment (e.g., precision/recall against manually verified Radar points or a manual verification study).
  3. [§III.A.1] For the night sequence, ground truth is produced by manually annotating restored RGB images produced by GSAD. The text does not specify whether these labels are applied to the raw or restored images; if applied to the raw night RGB images, the ground truth implicitly assumes that restoration preserves object boundaries and positions. No analysis of this label-transfer error is provided. The paper should either annotate the raw images directly or quantify the difference between annotations on restored and raw images, since night RGB is one of the benchmark modalities.
  4. [§IV] Despite the claim of a multi-modal benchmark, Section IV evaluates only RGB and TIR detection and tracking. There are no benchmark experiments using the LiDAR point-wise labels or the Radar point-wise labels, so the utility of those annotations for algorithm evaluation is not demonstrated. Adding at least basic LiDAR/Radar baseline experiments, or explicitly stating that these modalities are provided without benchmark results, would strengthen the paper.
  5. [§III, Fig. 2] The paper reports only approximate totals ('approximately 360,000 images' and 'around 190,000 labels') and a qualitative bar chart. For a benchmark dataset, exact per-sequence and per-modality label counts, class distributions, object-size distributions, and the exact train/validation/test splits are needed for reproducibility. These numbers are currently missing.
minor comments (5)
  1. [§II] Reference [2] is called the 'USVland dataset' in the text, but Table I and the reference title say 'USVInland'; please unify the name.
  2. [§III.B.1] The sentence 'To briefly explain the two transformations' is followed by only one equation; the text should be rephrased to match the content.
  3. [Fig. 2 caption] The caption says the vertical axis represents sequences, but the axes are not labeled in the figure, making it hard to read the exact label counts.
  4. [§II] The sentence 'we propose a PoLaRIS presented a multi-modal 3D object’s bounding box' is ungrammatical and should be rewritten.
  5. [Table II] For TIR rows, the Pre-train column uses '-' for COCO; the text explains this, but a footnote in the table would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the dataset labels are human-verified assets and the benchmark results are independent evaluations, not predictions derived from fitted inputs.

full rationale

The paper's chain is: construct multi-modal ground truth, then evaluate detectors and trackers against that fixed ground truth. Left RGB labels are initialized with YOLOv8 proposals but are then manually refined and manually completed for small objects, so the final labels are not defined by the detector. TIR labels are projected from left labels via Eq. (2) but the paper states that 'post-processing on the reference label is necessary' and that manual annotations are applied, making human verification the final arbiter. LiDAR labels are filtered by bounding-box projection and then manually verified ('we manually verify and remove points within the bounding box that do not belong to the object'), and Radar labels are obtained by clustering and overlap with the verified LiDAR labels. None of these steps fits a parameter and then renames that fit as a prediction; the benchmark metrics in Tables II and III are standard measurements on held-out test splits against this ground truth. The reuse of the Pohang Canal dataset [1] is an acknowledged external data source from another group, not a load-bearing self-citation. The geometric validity of Eq. (2), which omits per-object depth when transferring boxes between cameras with a translation, is a correctness or annotation-quality concern, not a circularity pattern, because the stated pipeline relies on manual correction rather than on the projection alone.

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

The central claim rests on the accuracy of the multi-modal annotations. The paper does not provide independent validation of the annotation quality, such as comparison to manually annotated subsets or error analysis of the sensor alignment.

free parameters (4)
  • DBSCAN clustering parameters = not specified
    Radar point-wise annotations are generated by clustering Radar data with DBSCAN (Section III.B.3). The eps and minPts values are not stated, so the annotation depends on undisclosed hyperparameters.
  • IoU filtering threshold = 0.8
    Overlapping initial bounding boxes are removed if IoU >= 0.8 (Eq. 1, Section III.A.1). This is a hand-set threshold that influences which annotations are kept.
  • Small-object size threshold = 5% of image area
    Small objects are defined as less than 5% of total image size (Section III.A.1). This determines what counts as a small object and which objects are manually annotated.
  • Training sample ratio = 5% of images per region
    Detection benchmarks sample 5% of images from each region for training (Section IV). This is a hand-set hyperparameter, not fitted.
assumptions (4)
  • domain assumption Extrinsic calibration between RGB, TIR, LiDAR, and Radar sensors is accurate
    The annotation pipeline projects labels between sensors using extrinsic parameters (Sections III.B.1 and III.B.2). No calibration error analysis is provided.
  • ad hoc to paper The Pohang Canal dataset provides accurate raw sensor data
    The entire dataset is built on the Pohang Canal dataset [1]; any sensor noise or synchronization error propagates to the annotations.
  • domain assumption Manual annotation and verification are correct and consistent
    The ground truth depends on human annotation and manual removal of false positives (Sections III.A and III.B), with no inter-annotator agreement measured.
  • domain assumption Diffusion-based GSAD restoration preserves object locations for night-time annotation
    Night RGB annotations are made on GSAD-enhanced images (Section III.A.1); if restoration distorts objects, the labels are wrong.

how reviews work

0 comments
Cite this review

Pith. "Pith review of PoLaRIS Dataset: A Maritime Object Detection and Tracking Dataset in Pohang Canal." pith.science (2026). https://pith.science/paper/A5GUBBN4

@misc{pith2026241206192,
  author       = {Pith},
  title        = {Pith review of: PoLaRIS Dataset: A Maritime Object Detection and Tracking Dataset in Pohang Canal},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A5GUBBN4}},
  note         = {Machine review of arXiv:2412.06192}
}
abstract

Maritime environments often present hazardous situations due to factors such as moving ships or buoys, which become obstacles under the influence of waves. In such challenging conditions, the ability to detect and track potentially hazardous objects is critical for the safe navigation of marine robots. To address the scarcity of comprehensive datasets capturing these dynamic scenarios, we introduce a new multi-modal dataset that includes image and point-wise annotations of maritime hazards. Our dataset provides detailed ground truth for obstacle detection and tracking, including objects as small as 10$\times$10 pixels, which are crucial for maritime safety. To validate the dataset's effectiveness as a reliable benchmark, we conducted evaluations using various methodologies, including \ac{SOTA} techniques for object detection and tracking. These evaluations are expected to contribute to performance improvements, particularly in the complex maritime environment. To the best of our knowledge, this is the first dataset offering multi-modal annotations specifically tailored to maritime environments. Our dataset is available at https://sites.google.com/view/polaris-dataset.

Figures

Figures reproduced from arXiv: 2412.06192 by the authors.

Figure 1
Figure 1. Pohang00 sequence’s 11194th scene. A clustered sparse Radar point cloud is scattered far and wide in the background. A dense red LiDAR point cloud is also scattered relatively short. To convert this scene PoLaRIS00, we first annotate a RGB image. Then, annotated bounding boxes in the RGB image are mapped to TIR. Finally, we extract only the LiDAR and Radar point clouds projected on annotated bounding boxes of the RG… view at source ↗
Figure 2
Figure 2. The vertical axis represents the Pohang00-04 sequences for the sensor modalities: image, LiDAR, and Radar. The horizontal axis indicates the number of labeled data for each sensor modality. Camera refers to left, right, and TIR image data, while Radar and LiDAR primarily detect dynamic obstacles and have limitations in identifying distant objects, resulting in significantly fewer data points compared to image data. … view at source ↗
Figure 3
Figure 3. The process of annotating the left image. (a) shows the annotation process during the day, where objects are more easily detected and labeled. [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: The process of semi-automatic annotation for multi-modal sensors. (a) shows the process of defining labels in the TIR image using transformation [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Illustration of dynamic objects. the moment of the stage of object detection, even for dynamic objects as small as 10×10 pixels. By providing tracking ground truth data for small, distant, and obscured objects, our dataset can validate experiments on objects of various…
Figure 6
Figure 6. Figure 6: File structure of our dataset. All label files for the image sensors are stored in a folder named all. The image labels and point-wise labels of dynamic objects are stored in the dynamic folder, as shown in [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]
Figure 7
Figure 7. Figure 7: The Pohang dataset [1] trajectory is shown, divided into four [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. MOANA: Multi-Radar Dataset for Maritime Odometry and Autonomous Navigation Application

    cs.RO 2024-12 conditional novelty 6.0 of 10

    MOANA provides seven maritime sequences with synchronized X-band and W-band radar, LiDAR, stereo images, GNSS, and radar/camera object labels, plus odometry benchmarks.

  2. MVTD: A Benchmark Dataset for Maritime Visual Object Tracking

    cs.CV 2025-06 conditional novelty 5.0 of 10

    MVTD is a 182-sequence, 150,000-frame maritime visual tracking benchmark with four object classes, showing that state-of-the-art trackers degrade on maritime scenes and improve after fine-tuning.

Reference graph

Works this paper leans on

35 extracted references · 21 canonical work pages · cited by 2 Pith papers

  1. [1]

    Pohang canal dataset: A multimodal mar- itime dataset for autonomous navigation in restricted waters,

    D. Chung, J. Kim, C. Lee, and J. Kim, “Pohang canal dataset: A multimodal mar- itime dataset for autonomous navigation in restricted waters,” The International Journal of Robotics Research , vol. 42, no. 12, pp. 1104–1114, 2023

    work page 2023

  2. [2]

    Are we ready for unmanned surface vehicles in inland waterways? the usvinland multisensor dataset and benchmark,

    Y . Cheng, M. Jiang, J. Zhu, and Y . Liu, “Are we ready for unmanned surface vehicles in inland waterways? the usvinland multisensor dataset and benchmark,” IEEE Robotics and Automation Letters , vol. 6, no. 2, pp. 3964–3970, 2021

    2021

  3. [3]

    Efficient obstacle detection based on prior estimation network and spatially constrained mixture model for unmanned surface vehicles,

    J. Liu, H. Li, J. Luo, S. Xie, and Y . Sun, “Efficient obstacle detection based on prior estimation network and spatially constrained mixture model for unmanned surface vehicles,” Journal of Field Robotics , vol. 38, no. 2, pp. 212–228, 2021

    work page 2021

  4. [4]

    KOLOMVERSE: Korea open large-scale image dataset for object detection in the maritime universe

    A. Nanda, S. W. Cho, H. Lee, and J. H. Park, “Kolomverse: Kriso open large- scale image dataset for object detection in the maritime universe,” arXiv preprint arXiv:2206.09885, 2022

    work page Pith review arXiv 2022

  5. [5]

    Video processing from electro-optical sensors for object detection and tracking in a maritime environment: A survey,

    D. K. Prasad, D. Rajan, L. Rachmawati, E. Rajabally, and C. Quek, “Video processing from electro-optical sensors for object detection and tracking in a maritime environment: A survey,” IEEE Transactions on Intelligent Transporta- tion Systems, vol. 18, no. 8, pp. 1993–2016, 2017

    work page 1993

  6. [6]

    Flow: A dataset and benchmark for floating waste detection in inland waters,

    Y . Cheng, J. Zhu, M. Jiang, J. Fu, C. Pang, P. Wang, K. Sankaran, O. Onabola, Y . Liu, D. Liu, and Y . Bengio, “Flow: A dataset and benchmark for floating waste detection in inland waters,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , October 2021, pp. 10 953–10 962

    work page 2021

  7. [7]

    The mastr1325 dataset for training deep usv obstacle detection models,

    B. Bovcon, J. Muhovi ˇc, J. Per ˇs, and M. Kristan, “The mastr1325 dataset for training deep usv obstacle detection models,” in 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2019, pp. 3431– 3438

    work page 2019

  8. [8]

    Analysis of different tracking algorithms applied on thermal infrared imagery for maritime surveillance systems,

    A. A. AlMansoori, I. Swamidoss, S. Sayadi, and A. Almarzooqi, “Analysis of different tracking algorithms applied on thermal infrared imagery for maritime surveillance systems,” in Artificial Intelligence and Machine Learning in Defense Applications II, vol. 11543. SPIE, 2020, pp. 30–40

    work page 2020

Show all 35 references
  1. [9]

    Vais: A dataset for recognizing maritime imagery in the visible and infrared spectrums,

    M. M. Zhang, J. Choi, K. Daniilidis, M. T. Wolf, and C. Kanan, “Vais: A dataset for recognizing maritime imagery in the visible and infrared spectrums,” in Proceedings of the IEEE conference on computer vision and pattern recognition workshops, 2015, pp. 10–16

    2015

  2. [10]

    Asynchronous trajectory matching-based multimodal maritime data fusion for vessel traffic surveillance in inland waterways,

    Y . Guo, R. W. Liu, J. Qu, Y . Lu, F. Zhu, and Y . Lv, “Asynchronous trajectory matching-based multimodal maritime data fusion for vessel traffic surveillance in inland waterways,” IEEE Transactions on Intelligent Transportation Systems , vol. 24, no. 11, pp. 12 779–12 792, 2023

    2023

  3. [11]

    Are we ready for autonomous driving? the kitti vision benchmark suite,

    A. Geiger, P. Lenz, and R. Urtasun, “Are we ready for autonomous driving? the kitti vision benchmark suite,” in 2012 IEEE conference on computer vision and pattern recognition. IEEE, 2012, pp. 3354–3361

    2012

  4. [12]

    Helipr: Heterogeneous lidar dataset for inter-lidar place recognition under spatiotemporal variations,

    M. Jung, W. Yang, D. Lee, H. Gil, G. Kim, and A. Kim, “Helipr: Heterogeneous lidar dataset for inter-lidar place recognition under spatiotemporal variations,” The International Journal of Robotics Research , p. 02783649241242136, 2023

    2023

  5. [13]

    Semantickitti: A dataset for semantic scene understanding of lidar sequences,

    J. Behley, M. Garbade, A. Milioto, J. Quenzel, S. Behnke, C. Stachniss, and J. Gall, “Semantickitti: A dataset for semantic scene understanding of lidar sequences,” in Proceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 9297–9307

    2019

  6. [14]

    Helimos: A dataset for moving object segmentation in 3d point clouds from heterogeneous lidar sensors,

    H. Lim, S. Jang, B. Mersch, J. Behley, H. Myung, and C. Stachniss, “Helimos: A dataset for moving object segmentation in 3d point clouds from heterogeneous lidar sensors,” arXiv preprint arXiv:2408.06328 , 2024

    2024 arXiv

  7. [15]

    Ultralytics yolov8,

    G. Jocher, A. Chaurasia, and J. Qiu, “Ultralytics yolov8,” 2023, version 8.0.0. [Online]. Available: https://github.com/ultralytics/ultralytics

    2023

  8. [16]

    Thermal object detection in difficult weather conditions using yolo,

    M. Kri ˇsto, M. Ivasic-Kos, and M. Pobar, “Thermal object detection in difficult weather conditions using yolo,” IEEE access, vol. 8, pp. 125 459–125 476, 2020

    2020

  9. [17]

    Global structure- aware diffusion process for low-light image enhancement,

    J. Hou, Z. Zhu, J. Hou, H. Liu, H. Zeng, and H. Yuan, “Global structure- aware diffusion process for low-light image enhancement,” Advances in Neural Information Processing Systems , vol. 36, 2024

    2024

  10. [18]

    Fieldscale: Locality-aware field-based adaptive rescaling for thermal infrared image,

    H. Gil, M.-H. Jeon, and A. Kim, “Fieldscale: Locality-aware field-based adaptive rescaling for thermal infrared image,” IEEE Robotics and Automation Letters , 2024

    2024

  11. [19]

    A. G. Bole, A. D. Wall, and A. Norris, Radar and ARPA manual: radar, AIS and target tracking for marine radar users . Butterworth-Heinemann, 2013

    2013

  12. [20]

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

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

    1996

  13. [21]

    Evaluating multiple object tracking perfor- mance: the clear mot metrics,

    K. Bernardin and R. Stiefelhagen, “Evaluating multiple object tracking perfor- mance: the clear mot metrics,”EURASIP Journal on Image and Video Processing, vol. 2008, pp. 1–10, 2008

    2008

  14. [22]

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

    E. Ristani, F. Solera, R. Zou, R. Cucchiara, and C. Tomasi, “Performance measures and a data set for multi-target, multi-camera tracking,” in European conference on computer vision . Springer, 2016, pp. 17–35

    2016

  15. [23]

    Ob- jects365: A large-scale, high-quality dataset for object detection,

    S. Shao, Z. Li, T. Zhang, C. Peng, G. Yu, X. Zhang, J. Li, and J. Sun, “Ob- jects365: A large-scale, high-quality dataset for object detection,” in Proceedings of the IEEE/CVF international conference on computer vision , 2019, pp. 8430– 8439

    2019

  16. [24]

    Gqa: A new dataset for real-world visual reasoning and compositional question answering,

    D. A. Hudson and C. D. Manning, “Gqa: A new dataset for real-world visual reasoning and compositional question answering,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2019, pp. 6700–6709

    2019

  17. [25]

    Flickr30k entities: Collecting region-to-phrase correspondences for richer image-to-sentence models,

    B. A. Plummer, L. Wang, C. M. Cervantes, J. C. Caicedo, J. Hockenmaier, and S. Lazebnik, “Flickr30k entities: Collecting region-to-phrase correspondences for richer image-to-sentence models,” in Proceedings of the IEEE international conference on computer vision , 2015, pp. 2641–2649

    2015

  18. [26]

    Grounding dino: Marrying dino with grounded pre-training for open-set object detection,

    S. Liu, Z. Zeng, T. Ren, F. Li, H. Zhang, J. Yang, C. Li, J. Yang, H. Su, J. Zhu et al., “Grounding dino: Marrying dino with grounded pre-training for open-set object detection,” arXiv preprint arXiv:2303.05499 , 2023

    2023 arXiv

  19. [27]

    Decoupled weight decay regularization,

    I. Loshchilov, “Decoupled weight decay regularization,” arXiv preprint arXiv:1711.05101, 2017

    2017 arXiv

  20. [28]

    Yolov10: Real-time end-to-end object detection,

    A. Wang, H. Chen, L. Liu, K. Chen, Z. Lin, J. Han, and G. Ding, “Yolov10: Real-time end-to-end object detection,” arXiv preprint arXiv:2405.14458 , 2024

    2024 arXiv

  21. [29]

    Detrs beat yolos on real-time object detection,

    Y . Zhao, W. Lv, S. Xu, J. Wei, G. Wang, Q. Dang, Y . Liu, and J. Chen, “Detrs beat yolos on real-time object detection,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024, pp. 16 965–16 974

    2024

  22. [30]

    Yolo-world: Real-time open-vocabulary object detection,

    T. Cheng, L. Song, Y . Ge, W. Liu, X. Wang, and Y . Shan, “Yolo-world: Real-time open-vocabulary object detection,” arXiv preprint arXiv:2401.17270 , 2024

    2024 arXiv

  23. [31]

    Microsoft coco: Common objects in context,

    T.-Y . Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Doll ´ar, and C. L. Zitnick, “Microsoft coco: Common objects in context,” in Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13 . Springer,...

    2014

  24. [32]

    Simple online and realtime tracking,

    A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, “Simple online and realtime tracking,” in 2016 IEEE international conference on image processing (ICIP) . IEEE, 2016, pp. 3464–3468

    2016

  25. [33]

    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,” in European conference on computer vision . Springer, 2022, pp. 1–21

    2022

  26. [34]

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

    J. Cao, J. Pang, X. Weng, R. Khirodkar, and K. Kitani, “Observation-centric sort: Rethinking sort for robust multi-object tracking,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2023, pp. 9686–9696

    2023

  27. [35]

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

    M. Yang, G. Han, B. Yan, W. Zhang, J. Qi, H. Lu, and D. Wang, “Hybrid-sort: Weak cues matter for online multi-object tracking,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 38, no. 7, 2024, pp. 6504–6512

    2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.