Pith. sign in

REVIEW 3 major objections 6 minor 2 cited by

MCBLT: Multi-Camera Multi-Object 3D Tracking in Long Videos

T0 review · 3 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 MCBLT claims multi-camera 3D tracking that holds identities across entire long videos, reaching 81.22 HOTA on AICity'24 and 95.6 IDF1 on WildTrack.

desk verdict Solid SOTA tracking paper with a genuinely useful untrained global merging block, but the priority claim over EarlyBird and the one-dataset support for the global block need referee attention. read the letter →

arxiv 2412.00692 v3 pith:D5QS42NT submitted 2024-12-01 cs.CV

Yizhou Wang , Tim Meinhardt , Orcun Cetintas , Cheng-Yen Yang , Sameer Satish Pusegaonkar , Benjamin Missaoui , Sujit Biswas , Zheng Tang
show 1 more author
Laura Leal-Taixé
This is my paper · ORCID
classification cs.CV
keywords multi-cameramulti-objecttrackingbird's-eyeviewdetectiongraphneuralnetworkslong-term3Dobjectre-identificationBEVFormertracking-by-detection
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 multi-camera 3D tracking of people in long videos is best done by merging all camera views into a single bird's-eye-view (BEV) representation before any detection, and by replacing handcrafted track-stitching heuristics with a global graph block that requires no extra training. The system, MCBLT, reaches 81.22 HOTA on the AICity'24 benchmark and 95.6 IDF1 on WildTrack, outperforming prior multi-camera trackers by clear margins. If the claim holds, surveillance and indoor monitoring in warehouses, retail, and hospitals would no longer lose identity every time a person disappears behind an obstacle or leaves one camera's view.

What carries the argument

The pipeline's load-bearing pieces are: (1) a BEVFormer spatio-temporal transformer adapted for static multi-camera scenes—spatial cross-attention projects BEV reference points into each camera view, temporal self-attention folds in past BEV features, and a DETR head regresses 3D boxes; (2) a 2D-3D detection association algorithm that projects each 3D box into each view and Hungarian-matches it to 2D detections using bottom-center distance with an occlusion-aware penalty, producing clean per-person ReID crops; and (3) a hierarchical GNN (SUSHI extended to 3D) whose nodes are detections and tracklets and whose edges are classification hypotheses, crowned by a global block that connects incoming tracklets to past tracks across the whole video while sharing weights with the previous hierarchy level. The global block is what enables association beyond the sliding-window size without handcrafted stitching heuristics.

What would settle it

Take a long AICity'24 sequence, keep detections fixed, and artificially enlarge every occlusion gap so that each gap is longer than the largest sliding-window size used in the ablation; if the untrained global block does not keep HOTA roughly flat compared to the short-window version, then the claimed full-video association does not come from the shared-weight transfer.

Watch

Extended reading notes

Core claim

MCBLT is the first 3D multi-target multi-camera tracker to combine early multi-view BEV aggregation with a hierarchical graph neural network. Detections are produced by an adapted BEVFormer that fuses calibrated camera images into BEV features; each 3D box is then matched to the best 2D detection per view so that ReID features come from tight person crops rather than oversized projected boxes. Tracking happens in 3D space with a hierarchical GNN whose final 'global block' shares weights with the previous hierarchy level and therefore needs no training, letting the tracker associate objects across the full length of a video—up to 24,000 frames with occlusions up to 2,000 frames. On the AICity'24 test set the method reports 81.22 HOTA, +9.28 over the best prior submission, and on WildTrack it reports 95.6 IDF1 using the same detections as the EarlyBird baseline, +3.3 over it.

Load-bearing premise

The global block works by sharing weights with the previous hierarchy level, so the edge classifier trained on detection-to-detection links inside a sliding window must also correctly judge tracklet-to-tracklet links across the whole video.

Editorial extensions

If this is right

  • Long videos with occlusion gaps up to 2,000 frames no longer require hand-tuned window stitching; the global block associates tracks over the full clip.
  • New scenes with different camera counts and placements can reuse the same detection and tracking models, since the BEVFormer adaptation removes camera embeddings and re-centers the BEV origin.
  • Because the global block needs no training, the tracker can be applied to arbitrarily long sequences without new annotations or fine-tuning.
  • The 2D-3D association step alone lifts WildTrack IDF1 from 63.2 to 93.4, showing that ReID feature quality is a primary bottleneck for MTMC tracking.

Reading between the lines

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

  • Editorial: the untrained global block is a form of self-distillation, reusing the edge classifier learned on short detection edges across a much longer time scale; a stress test that artificially breaks temporal gaps beyond the training window would reveal whether long-gap identity switches concentrate there.
  • Editorial: the same early-fusion-plus-hierarchical-GNN recipe should transfer to other object classes and to drone or vehicle fleets where calibrated 3D positions exist, though appearance features would need to be class-specific.
  • Editorial: because the ReID features are trained partly on synthetic AICity'24 data and evaluated on real WildTrack, the large IDF1 gap suggests that domain-adapting the ReID head could improve real-world performance more than further architectural changes.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper presents MCBLT, a multi-camera 3D detection and tracking framework. It adapts BEVFormer to static MTMC scenes for early multi-view BEV detection, extracts ReID features by associating 3D detections with 2D detections, and performs 3D multi-object tracking via a hierarchical GNN inspired by SUSHI. A novel global tracking block stitches non-overlapping windows without additional training. On the AICity'24 test set the paper reports 81.22 HOTA, a +9.28 improvement over the best prior submission; on WildTrack it reports 95.6 IDF1 when using the same detections as EarlyBird, an improvement of +3.3 IDF1 and +3.1 MOTA. Ablations attribute large gains to the global block and to the 2D-3D detection association algorithm.

Significance. If the reported numbers are reproducible, this is a strong empirical contribution: it is one of the first 3D hierarchical GNN trackers for MTMC, and the clean same-detection comparison (MCBLT† vs. EarlyBird) and the same-weights ablation in Table 5 directly compare heuristic stitching with the learned global block. The paper also provides internal consistency checks and supplementary analyses of ReID quality and runtime. However, the two headline claims—long-term association and generalizability—rest on assumptions that are not directly tested, and the absence of code/model release and of GNN training details weakens verifiability.

major comments (3)
  1. [3.4; Table 5] The claim that the untrained global block 'unlocks occlusion handling beyond the number of frames per graph' rests on the assumption that an edge classifier trained on within-window detection-to-detection edges also classifies full-video tracklet-to-tracklet edges correctly. Table 5 shows only aggregate HOTA/AssA gains on one dataset, and the comparison changes both the association span and the stitching mechanism at once, so it does not isolate the validity of weight sharing. The manuscript should report performance stratified by occlusion-gap length (e.g., HOTA/AssA for gaps of 100, 500, 1000, and 2000 frames), compare edge-feature distributions between within-window and cross-window edges, and include an ablation that replaces the shared-weight global block with a separately trained or fine-tuned global block. Without this, the long-term association claim is not supported beyond a single aggregate benchmark.
  2. [4.2; Appendix B.1] The manuscript does not provide training details for the core tracking model: no paragraph in Section 4.2 describes the SUSHI/GNN training procedure, loss, optimizer, epochs, hierarchy configuration, window size, or graph-pruning thresholds. In addition, no code or model is released, several thresholds (α in Eq. 5, the IoU gate in Eq. 4, and δ and Δ in Algorithm 1) are either not reported or reported without sensitivity analysis, and no error bars are given. These omissions make it impossible to independently verify the SOTA numbers or to determine whether the reported gains are robust to hyperparameter choices. The paper should release code/models or provide a complete public-data implementation together with a sensitivity analysis.
  3. [4.4; Appendix A.3; Abstract] The abstract and contributions claim 'impressive generalizability across different scenes and diverse camera settings,' but the experimental support is indirect. The WildTrack result is obtained after fine-tuning, with pre-training on AICity'24, not by zero-shot transfer to an unseen scene or camera configuration, and WildTrack is a single scene. To support the generalizability claim, the paper should include an experiment that trains on AICity'24 and evaluates on WildTrack (or another scene) without fine-tuning, or systematically varies the number of cameras and camera placements at inference time. As written, the claim exceeds the evidence.
minor comments (6)
  1. [Section 1, contribution bullet] The bullet 'The first MTMC method efficiently performs early multi-view image aggregation' is inconsistent with EarlyBird [30], which is cited as an early-fusion method; rephrase to claim the first combination of early multi-view aggregation with hierarchical GNN tracking.
  2. [Eq. (5)] The value of α is used in the cost function but never reported; add the value and a brief sensitivity check.
  3. [Table 2] The 'detection association accuracy' metric is introduced informally; define it formally in the main text, and correct the apparent typo 'corrected matches' to 'correct matches'.
  4. [Table 3] MCBLT† has higher IDF1 and MOTA than EarlyBird but a worse ML value (7.3 vs. 4.9); the paper should briefly explain why the mostly-lost rate increases despite better association.
  5. [Appendix C] The capitalization 'Wildtrack' is inconsistent with 'WildTrack' used elsewhere; use the official capitalization throughout.
  6. [Table 11] The runtime table is confusingly laid out; clarify that DINO, SOLIDER, and SUSHI can run in parallel with BEVFormer, and state explicitly which components are included in the reported 1.5 FPS end-to-end number.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the derivation chain is benchmark-driven and externally evaluated.

full rationale

The paper's central results are external benchmark scores on AICity'24 and WildTrack, not quantities derived from the method's own definitions. The tracking module is adapted from the authors' prior SUSHI work, but SUSHI is treated as a concrete baseline and implementation to build on and ablate against (Tables 5 and 6), not as an unverified theorem that forces the conclusion. The untrained global block in Section 3.4 shares weights with the previous hierarchy level, which is a transfer assumption rather than a circular reduction: the paper does not define the global block's output as the training target, and Table 5 compares it against SUSHI's heuristic stitching under otherwise similar conditions. The ReID and detector components are evaluated on benchmark splits and with independent comparisons such as MCBLT using EarlyBird's detections. No equation in the paper is fit to the reported HOTA or IDF1 values, and no claimed prediction reduces by construction to a fitted parameter or to a self-citation chain.

Assumptions & free parameters 8 free parameters · 6 assumptions · 0 invented entities

The central performance claim is empirical, not a derivation. It rests on calibrated projection, a set of hand-set thresholds, dataset-specific assumptions about people, and one untested distribution-shift assumption for the global block. No new physical entities are introduced; the global block is a method component whose independent evidence is the ablation in Table 5.

free parameters (8)
  • CircleNMS threshold delta = 0.2 m
    Filters nearby false positives in BEV detections; set by hand in Section 4.2 and Appendix B.1 without a published sensitivity study.
  • Hungarian cost threshold Delta = 150
    Limits accepted 2D-3D matches; set in Appendix B.1 and not ablated.
  • IoU gate for 2D-3D association = 0.1
    Equation (4) rejects matches below this overlap; hand-set without sensitivity analysis.
  • Occlusion penalty alpha = >1 (value not stated)
    Equation (5) penalizes 2D boxes lower than 3D projections; exact value is omitted.
  • Default person box dimension on WildTrack = [0.6, 0.6, 1.7] m, zero rotation
    WildTrack has no 3D box annotations; fixed dimensions are used for projection and association in Section 4.2.
  • BEV resolution and voxel size = 200x200, 0.5 m
    Chosen by the validation mAP sweep in Table 4.
  • Training epochs and learning rate = 48 epochs, 2e-4
    Selected for detection convergence in Table 4 and Appendix A.1.
  • GNN sliding-window stride s = not reported
    Controls the overlap-free near-online processing in Figure 4; the value is not stated.
assumptions (6)
  • domain assumption Known camera intrinsics, extrinsics, and ground-plane world coordinates are available for every scene, with projection given by Equation (1).
    The whole method depends on calibrated multi-view geometry; Section 3.1 defines the coordinate systems.
  • domain assumption The world origin can be re-centered to the scene floor-plan center and the ground plane is z=0.
    Appendix A.2 shows re-centering improves mAP by +22.33, indicating the method is sensitive to this assumption.
  • domain assumption People can be represented as 3D boxes with fixed default dimensions [0.6, 0.6, 1.7] m and zero rotation on WildTrack.
    Section 4.2 makes this assumption because WildTrack has no 3D box annotations.
  • domain assumption Synthetic AICity'24 data transfers to the real WildTrack scene through pre-training.
    Appendix A.3 reports +3.67 mAP from pre-training, but transferability is assumed beyond this single real-world scene.
  • ad hoc to paper The edge classifier trained on within-window detection edges also applies to cross-window tracklet edges in the untrained global block.
    Section 3.4 states the global block shares weights and requires no additional training, but no distribution-shift analysis is given.
  • standard math Standard graph message passing, pruning, and Hungarian assignment preserve network-flow integrity and identity consistency.
    Equations (6)-(8) and the graph formulation follow prior work [3, 5, 42] and are treated as accepted algorithmic machinery.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MCBLT: Multi-Camera Multi-Object 3D Tracking in Long Videos." pith.science (2026). https://pith.science/paper/D5QS42NT

@misc{pith2026241200692,
  author       = {Pith},
  title        = {Pith review of: MCBLT: Multi-Camera Multi-Object 3D Tracking in Long Videos},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/D5QS42NT}},
  note         = {Machine review of arXiv:2412.00692}
}
abstract

Object perception from multi-view cameras is crucial for intelligent systems, particularly in indoor environments, e.g., warehouses, retail stores, and hospitals. Most traditional multi-target multi-camera (MTMC) detection and tracking methods rely on 2D object detection, single-view multi-object tracking (MOT), and cross-view re-identification (ReID) techniques, without properly handling important 3D information by multi-view image aggregation. In this paper, we propose a 3D object detection and tracking framework, named MCBLT, which first aggregates multi-view images with necessary camera calibration parameters to obtain 3D object detections in bird's-eye view (BEV). Then, we introduce hierarchical graph neural networks (GNNs) to track these 3D detections in BEV for MTMC tracking results. Unlike existing methods, MCBLT has impressive generalizability across different scenes and diverse camera settings, with exceptional capability for long-term association handling. As a result, our proposed MCBLT establishes a new state-of-the-art on the AICity'24 dataset with $81.22$ HOTA, and on the WildTrack dataset with $95.6$ IDF1.

Figures

Figures reproduced from arXiv: 2412.00692 by the authors.

Figure 1
Figure 1. Comparison among three types of MTMC tracking [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overall framework of MCBLT. First, multi-view images at frame [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Illustration on 2D detection (in blue) and the correspond [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: GNN hierarchy of MCBLT for tracking. We process [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Visualization of MTMC detection and tracking results for three different scenes in AICity’24 test set. The tracked objects are [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Visualization of MTMC detection and tracking results on WildTrack test set. The tracked objects are shown as colored poles [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Visualization of 2D-3D detection association results. [PITH_FULL_IMAGE:figures/full_fig_p012_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. MVTrajecter: Multi-View Pedestrian Tracking with Trajectory Motion Cost and Trajectory Appearance Cost

    cs.CV 2025-09 conditional novelty 6.0 of 10

    An end-to-end multi-view pedestrian tracker that aggregates motion and appearance costs over K past timestamps, outperforming prior methods on GMVD, Wildtrack, and MultiviewX, though with a validation-protocol concern...

  2. IndoorBEV: Joint Detection and Footprint Completion of Objects via Mask-based Prediction in Indoor Scenarios for Bird's-Eye View Perception

    cs.RO 2025-07 conditional novelty 4.0 of 10

    IndoorBEV uses a query-based transformer decoder on a bird's-eye view lidar grid to jointly detect objects and predict footprint masks in indoor scenes.

Reference graph

Works this paper leans on

47 extracted references · 37 canonical work pages · cited by 2 Pith papers

  1. [1]

    Deep oc- clusion reasoning for multi-camera multi-target detection

    Pierre Baqu ´e, Franc ¸ois Fleuret, and Pascal Fua. Deep oc- clusion reasoning for multi-camera multi-target detection. In Proceedings of the IEEE International Conference on Com- puter Vision, pages 271–279, 2017. 2

    2017

  2. [2]

    Evaluating mul- tiple object tracking performance: the clear mot metrics

    Keni Bernardin and Rainer Stiefelhagen. Evaluating mul- tiple object tracking performance: the clear mot metrics. EURASIP Journal on Image and Video Processing, 2008:1– 10, 2008. 7

    work page 2008

  3. [3]

    Learning a neural solver for multiple object tracking

    Guillem Bras ´o and Laura Leal-Taix ´e. Learning a neural solver for multiple object tracking. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 5

    work page 2020

  4. [4]

    Data association for multi-object tracking- by-detection in multi-camera networks

    Michael Bredereck, Xiaoyan Jiang, Marco K ¨orner, and Joachim Denzler. Data association for multi-object tracking- by-detection in multi-camera networks. In 2012 Sixth International Conference on Distributed Smart Cameras (ICDSC), pages 1–6. IEEE, 2012. 2

    work page 2012

  5. [5]

    Uni- fying short and long-term tracking with graph hierarchies

    Orcun Cetintas, Guillem Bras ´o, and Laura Leal-Taix´e. Uni- fying short and long-term tracking with graph hierarchies. In Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 22877–22887, 2023. 2, 4, 5, 8

    work page 2023

  6. [6]

    Wild- track: A multi-camera hd dataset for dense unscripted pedes- trian detection

    Tatjana Chavdarova, Pierre Baqu ´e, St ´ephane Bouquet, An- drii Maksai, Cijo Jose, Timur Bagautdinov, Louis Lettry, Pascal Fua, Luc Van Gool, and Franc ¸ois Fleuret. Wild- track: A multi-camera hd dataset for dense unscripted pedes- trian detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 5030–5039,

    work page

  7. [7]

    Beyond appearance: a semantic controllable self-supervised learning framework for human-centric visual tasks

    Weihua Chen, Xianzhe Xu, Jian Jia, Hao Luo, Yaohua Wang, Fan Wang, Rong Jin, and Xiuyu Sun. Beyond appearance: a semantic controllable self-supervised learning framework for human-centric visual tasks. In The IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, 2023. 6

    work page 2023

  8. [8]

    Rest: A reconfigurable spatial-temporal graph model for multi-camera multi-object tracking

    Cheng-Che Cheng, Min-Xuan Qiu, Chen-Kuo Chiang, and Shang-Hong Lai. Rest: A reconfigurable spatial-temporal graph model for multi-camera multi-object tracking. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 10051–10060, 2023. 1, 2, 6, 7

    work page 2023

Show all 47 references
  1. [9]

    Multi-camera object de- tection for robotics

    Adam Coates and Andrew Y Ng. Multi-camera object de- tection for robotics. In 2010 IEEE International conference on robotics and automation, pages 412–419. IEEE, 2010. 2

    2010

  2. [10]

    Homography based multiple camera detection and tracking of people in a dense crowd

    Ran Eshel and Yael Moses. Homography based multiple camera detection and tracking of people in a dense crowd. In 2008 IEEE Conference on Computer Vision and Pattern Recognition, pages 1–8. IEEE, 2008. 2

    2008

  3. [11]

    Multiview detection with shadow transformer (and view-coherent data augmentation)

    Yunzhong Hou and Liang Zheng. Multiview detection with shadow transformer (and view-coherent data augmentation). In Proceedings of the 29th ACM International Conference on Multimedia, pages 1673–1682, 2021. 2, 6, 7

    2021

  4. [12]

    Traffic-aware multi-camera tracking of vehicles based on reid and camera link model

    Hung-Min Hsu, Yizhou Wang, and Jenq-Neng Hwang. Traffic-aware multi-camera tracking of vehicles based on reid and camera link model. In Proceedings of the 28th ACM International Conference on Multimedia , pages 964– 972, 2020. 2

    2020

  5. [13]

    Multi-target multi-camera tracking of vehicles using metadata-aided re-id and trajectory-based camera link model

    Hung-Min Hsu, Jiarui Cai, Yizhou Wang, Jenq-Neng Hwang, and Kwang-Ju Kim. Multi-target multi-camera tracking of vehicles using metadata-aided re-id and trajectory-based camera link model. IEEE Transactions on Image Processing, 30:5198–5210, 2021. 2

    2021

  6. [14]

    Principal axis-based correspon- dence between multiple cameras for people tracking

    Weiming Hu, Min Hu, Xue Zhou, Tieniu Tan, Jianguang Lou, and Steve Maybank. Principal axis-based correspon- dence between multiple cameras for people tracking. IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(4):663–671, 2006. 2

    2006

  7. [15]

    A new approach to linear filtering and prediction problems

    Rudolph Emil Kalman. A new approach to linear filtering and prediction problems. 1960. 3

    1960

  8. [16]

    Cluster self-refinement for enhanced online multi- camera people tracking

    Jeongho Kim, Wooksu Shin, Hancheol Park, and Donghyuk Choi. Cluster self-refinement for enhanced online multi- camera people tracking. In CVPR Workshop, Seattle, W A, USA, 2024. 7

    2024

  9. [17]

    The open images dataset v4: Unified image classification, object detection, and visual relationship detection at scale

    Alina Kuznetsova, Hassan Rom, Neil Alldrin, Jasper Ui- jlings, Ivan Krasin, Jordi Pont-Tuset, Shahab Kamali, Stefan Popov, Matteo Malloci, Alexander Kolesnikov, et al. The open images dataset v4: Unified image classification, object detection, and visual relationship detection...

    1956

  10. [18]

    Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers

    Zhiqi Li, Wenhai Wang, Hongyang Li, Enze Xie, Chong- hao Sima, Tong Lu, Yu Qiao, and Jifeng Dai. Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers. In European con- ference on computer vision, pages 1–18. Springer, 2022...

    2022

  11. [19]

    Swin transformer: Hierarchical vision transformer using shifted windows

    Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo. Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021. 6

    2021

  12. [20]

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

    Jonathon Luiten, Aljosa Osep, Patrick Dendorfer, Philip Torr, Andreas Geiger, Laura Leal-Taix´e, and Bastian Leibe. Hota: A higher order metric for evaluating multi-object tracking. International Journal of Computer Vision , pages 1–31, 2020. 6

    2020

  13. [21]

    Mot16: A benchmark for multi-object tracking

    Anton Milan, Laura Leal-Taix ´e, Ian Reid, Stefan Roth, and Konrad Schindler. Mot16: A benchmark for multi-object tracking. arXiv preprint arXiv:1603.00831, 2016. 5

    2016 arXiv

  14. [22]

    Lmgp: Lifted mul- ticut meets geometry projections for multi-camera multi- object tracking

    Duy MH Nguyen, Roberto Henschel, Bodo Rosenhahn, Daniel Sonntag, and Paul Swoboda. Lmgp: Lifted mul- ticut meets geometry projections for multi-camera multi- object tracking. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 8866– 887...

    2022

  15. [23]

    Nvidia omniverse platform

    NVIDIA. Nvidia omniverse platform. https://www. nvidia . com / en - us / design - visualization / omniverse/, 2021. Accessed: Apr 10, 2023. 6

    2021

  16. [24]

    A bayesian filter for multi-view 3d multi- object tracking with occlusion handling

    Jonah Ong, Ba-Tuong V o, Ba-Ngu V o, Du Yong Kim, and Sven Nordholm. A bayesian filter for multi-view 3d multi- object tracking with occlusion handling. IEEE Transactions on Pattern Analysis and Machine Intelligence , 44(5):2246– 2263, 2020. 7

    2020

  17. [25]

    Performance measures and a data set for 9 multi-target, multi-camera tracking

    Ergys Ristani, Francesco Solera, Roger Zou, Rita Cucchiara, and Carlo Tomasi. Performance measures and a data set for 9 multi-target, multi-camera tracking. In European conference on computer vision, pages 17–35. Springer, 2016. 7

    2016

  18. [26]

    Conditional random fields for multi-camera object de- tection

    Gemma Roig, Xavier Boix, Horesh Ben Shitrit, and Pascal Fua. Conditional random fields for multi-camera object de- tection. In 2011 International Conference on Computer Vi- sion, pages 563–570. IEEE, 2011. 2

    2011

  19. [27]

    Object detection, tracking and recogni- tion for multiple smart cameras

    Aswin C Sankaranarayanan, Ashok Veeraraghavan, and Rama Chellappa. Object detection, tracking and recogni- tion for multiple smart cameras. Proceedings of the IEEE , 96(10):1606–1624, 2008. 2

    2008

  20. [28]

    Ocmctrack: Online multi-target multi- camera tracking with corrective matching cascade

    Andreas Specker. Ocmctrack: Online multi-target multi- camera tracking with corrective matching cascade. In CVPR Workshop, Seattle, W A, USA, 2024. 7

    2024

  21. [29]

    Online multi-camera people tracking with spatial-temporal mechanism and anchor-feature hierarchical clustering

    Vasin Suttichaya, Riu Cherdchusakulchai, Sasin Phimsiri, Visarut Trairattanapa, Suchat Tungjitnob, Wasu Kudis- thalert, Pornprom Kiawjak, Ek Thamwiwatthana, Phawat Borisuitsawat, Teepakorn Tosawadi, Pakcheera Choppra- dit, Kasisdis Mahakijdechachai, Supawit Vatathanavaro, and ...

    2024

  22. [30]

    Earlybird: Early-fusion for multi- view tracking in the bird’s eye view

    Torben Teepe, Philipp Wolters, Johannes Gilg, Fabian Her- zog, and Gerhard Rigoll. Earlybird: Early-fusion for multi- view tracking in the bird’s eye view. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 102–111, 2024. 1, 2, 6, 7, 8

    2024

  23. [31]

    Efficient online multi-camera tracking with memory-efficient accumulated appearance fea- tures and trajectory validation

    Huan Vi and Lap Quoc Tran. Efficient online multi-camera tracking with memory-efficient accumulated appearance fea- tures and trajectory validation. In CVPR Workshop, Seattle, W A, USA, 2024. 7

    2024

  24. [32]

    The 8th ai city challenge

    Shuo Wang, David C Anastasiu, Zheng Tang, Ming- Ching Chang, Yue Yao, Liang Zheng, Mohammed Shaiqur Rahman, Meenakshi S Arya, Anuj Sharma, Pranamesh Chakraborty, et al. The 8th ai city challenge. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recogn...

    2024

  25. [33]

    Towards real-time multi-object tracking

    Zhongdao Wang, Liang Zheng, Yixuan Liu, and Shengjin Wang. Towards real-time multi-object tracking. The Euro- pean Conference on Computer Vision (ECCV), 2020. 8

    2020

  26. [34]

    A robust online multi- camera people tracking system with geometric consistency and state-aware re-id correction

    Zhenyu Xie, Zelin Ni, Wenjie Yang, Yuang Zhang, Yihang Chen, Yang Zhang, and Xiao Ma. A robust online multi- camera people tracking system with geometric consistency and state-aware re-id correction. InCVPR Workshop, Seattle, W A, USA, 2024. 7

    2024

  27. [35]

    Multi-view people tracking via hierarchical trajectory com- position

    Yuanlu Xu, Xiaobai Liu, Yang Liu, and Song-Chun Zhu. Multi-view people tracking via hierarchical trajectory com- position. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4256–4265, 2016. 1, 2

    2016

  28. [36]

    Cross-view people tracking by scene-centered spatio- temporal parsing

    Yuanlu Xu, Xiaobai Liu, Lei Qin, and Song-Chun Zhu. Cross-view people tracking by scene-centered spatio- temporal parsing. In Proceedings of the AAAI conference on artificial intelligence, 2017. 1, 2

    2017

  29. [37]

    Bevformer v2: Adapting modern image backbones to bird’s-eye-view recognition via perspective su- pervision

    Chenyu Yang, Yuntao Chen, Hao Tian, Chenxin Tao, Xizhou Zhu, Zhaoxiang Zhang, Gao Huang, Hongyang Li, Yu Qiao, Lewei Lu, et al. Bevformer v2: Adapting modern image backbones to bird’s-eye-view recognition via perspective su- pervision. In Proceedings of the IEEE/CVF Conference...

    2023

  30. [38]

    An online approach and evaluation method for tracking people across cameras in extremely long video sequence

    Cheng-Yen Yang, Hsiang-Wei Huang, Pyong-Kun Kim, Zhongyu Jiang, Kwang-Ju Kim, ChungI Huang, Haiqing Du, and Jenq-Neng Hwang. An online approach and evaluation method for tracking people across cameras in extremely long video sequence. In CVPR Workshop, Seattle, W A, USA,

  31. [39]

    Overlap suppression clustering for offline multi-camera people tracking

    Ryuto Yoshida, Junichi Okubo, Junichiro Fujii, Masazumi Amakata, and Takayoshi Yamashita. Overlap suppression clustering for offline multi-camera people tracking. InCVPR Workshop, Seattle, W A, USA, 2024. 7

    2024

  32. [40]

    Real-time 3d deep multi- camera tracking

    Quanzeng You and Hao Jiang. Real-time 3d deep multi- camera tracking. arXiv preprint arXiv:2003.11753 , 2020. 7

    2003 arXiv

  33. [41]

    Dino: Detr with improved denoising anchor boxes for end-to-end object detection

    Hao Zhang, Feng Li, Shilong Liu, Lei Zhang, Hang Su, Jun Zhu, Lionel M Ni, and Heung-Yeung Shum. Dino: Detr with improved denoising anchor boxes for end-to-end object detection. arXiv preprint arXiv:2203.03605, 2022. 6

    2022 arXiv

  34. [42]

    Global data as- sociation for multi-object tracking using network flows

    Li Zhang, Yuan Li, and Ramakant Nevatia. Global data as- sociation for multi-object tracking using network flows. In CVPR, 2008. 5

    2008

  35. [43]

    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, 129:3069–3087, 2021. 3

    2021

  36. [44]

    Scalable person re-identification: A benchmark

    Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jing- dong Wang, and Qi Tian. Scalable person re-identification: A benchmark. In Computer Vision, IEEE International Con- ference on, 2015. 6

    2015

  37. [45]

    Understanding the robustness in vision transformers

    Daquan Zhou, Zhiding Yu, Enze Xie, Chaowei Xiao, An- imashree Anandkumar, Jiashi Feng, and Jose M Alvarez. Understanding the robustness in vision transformers. In In- ternational Conference on Machine Learning, pages 27378– 27394. PMLR, 2022. 6

    2022

  38. [46]

    Ob- jects as points

    Xingyi Zhou, Dequan Wang, and Philipp Kr ¨ahenb¨uhl. Ob- jects as points. arXiv preprint arXiv:1904.07850, 2019. 3

    1904 arXiv

  39. [47]

    re-centering

    Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, and Jifeng Dai. Deformable detr: Deformable transformers for end-to-end object detection. In International Conference on Learning Representations, 2022. 2, 3 10 MCBLT: Multi-Camera Multi-Object 3D Tracking in Long Videos ...

    2022

Pith tools

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