Pith. sign in

REVIEW 3 major objections 4 minor 115 references

Reviewed by Pith at T0; open to challenge.

T0 means a machine referee read the full paper against a public rubric. The mark states how deep the mechanical check went, never who wrote it. the ladder, T0–T4 →

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

T0 review · glm-5.2

427 Teams, 5 Tasks: SoccerNet 2026 Benchmarks Sports Video Understanding

2026-07-09 14:08 UTC pith:PGO4B35D

load-bearing objection Solid challenge report; small test sets for BAA and VQA need acknowledgment the 3 major comments →

arxiv 2607.07320 v1 pith:PGO4B35D submitted 2026-07-08 cs.CV

SoccerNet 2026 Challenges Results

Anthony Cioppa , Silvio Giancola , H{\aa}kan Ard\"o , Mohamad Dalal , Jan Held , J\'er\'emie Ochin , Jiayuan Rao , Karen Sanchez
show 97 more authors
Renaud Vandeghen Artur Xarles Olivier Barnich Albert Clap\'es Mathieu Delvaux Sergio Escalera Bernard Ghanem C\'edric Hons Antoine Houet Sotiris Manitsaris Tom Michel Pierre Miralles Thomas B. Moeslund Mikael Nilsson Bogdan Stanciulescu Marc Van Droogenbroeck Yanfeng Wang Weidi Xie Faisal Altawijri Mohamed Atef Semen Budennyy Vasiliy Chelpanov Puhua Chen Yixin Chen Lechao Cheng Jianling Chu Ju-Seong Do Oleg Durygin Omar Fetouh Mirco Fuchs Youssef Ghallab Falguni Ghosh Wonjun Heo Yufeng Hu Weixuan Huang Phuong-Linh Huynh-Ha Matvey Isupov Yangguang Ji Siyuan Jiang Zhenxiang Jiang Wonyong Jo Ho-Young Jung SeongHeon Kang MinJae Kim Youngseon Kim Jakub Komosa Artem Konshin Trung-Hoang Le Jongmin Lee Lingling Li Litao Li Vadim Linkov Fang Liu Haoxuan Ma Shun Makino Ismail Mathkour Konstantin Mitin Mikhail Moiseev Takumi Nagaya Yuki Nakamura Thanh-Khoi Nguyen Hoang-Phuc Nguyen Trong-Thuan Nguyen Christian Orduz Kwanyong Park Fabian Perez Parthsarthi Rawat SuHyun Rim Hoover Rueda-Chac\'on Atom Scott Minori Sugimura Yuyang Sun Shengeng Tang Minh-Triet Tran Ikuma Uchida Juan Vanegas Thanh-Nhan Vo Jiangtao Wang Yaxiong Wang Xiaogang Wang Ruifeng Wang Rio Watanabe Jiali Wen Yongliang Wu Di Yang Xu Yang Zhuo Yang Xinyu Ye Yibo Yu Zihan Zhai Yu Zhang Zhenyu Zhao Zhun Zhong Yixi Zhou Xingyu Zhu Wenbo Zhu Julian Ziegler
This is my paper
classification cs.CV
keywords soccer video understandingaction anticipationaction spottingnovel view synthesisathlete localizationvisual question answeringbenchmarkcomputer vision
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper documents the sixth annual SoccerNet challenge, a community benchmark for computer vision in soccer video. Across five tasks—predicting near-future ball actions, attributing actions to individual players, synthesizing novel camera views, localizing athletes in world coordinates from static cameras, and answering natural-language questions about broadcasts—427 teams submitted 1,129 entries. The leading submission for every task improved over the provided baseline. The paper describes each task definition, dataset, evaluation protocol, leaderboard, and winning method, then identifies cross-task themes: higher input resolution, larger or ensembled models, careful confidence calibration, and explicit use of domain structure such as camera geometry and tactical game-state features consistently drove gains. The paper positions itself as a reference snapshot of the current state of the art in sports video understanding as measured on held-out challenge data.

Core claim

The central finding is that across five diverse soccer video understanding tasks, the performance gains that separated winning submissions from baselines came from a convergent set of engineering strategies rather than fundamentally new architectures. Higher input resolution captured fine-grained visual cues for small or distant players; model ensembles mitigated the uncertainty inherent in tasks like action anticipation and novel view synthesis; confidence calibration and class-imbalance mitigation addressed the long-tailed distribution of soccer actions; and explicit injection of domain knowledge—camera calibration geometry for athlete localization, tactical game-state features for action—

What carries the argument

The five benchmark tasks themselves are the central objects: Ball Action Anticipation (predicting action class and timing in an unobserved 5-second window from 30 seconds of video), Player-Centric Ball Action Spotting (localizing and classifying actions while assigning them to specific players via team and jersey number), Novel View Synthesis (rendering images from unobserved camera poses in multi-view soccer scenes), Spiideo SoccerNet Synloc (localizing athletes in real-world pitch coordinates from a single calibrated static-camera image), and Visual Question Answering (answering multiple-choice questions about soccer broadcasts across text, image, and video). Each task is paired with a专用数据

Load-bearing premise

The paper assumes that the challenge evaluation protocols and held-out test splits are sufficient to draw reliable conclusions about method performance, but the small test set sizes (for example, 2 matches for Ball Action Anticipation, 500 questions for VQA) introduce high variance into the rankings, meaning that the ordering of top teams may not be statistically robust.

What would settle it

If the cross-task themes (resolution, ensembling, calibration, domain structure) identified as drivers of improvement were not actually the load-bearing factors in the winning submissions, the paper's methodological synthesis would be unsupported. This could be tested by ablating each factor in isolation.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • The convergence on resolution, ensembling, and calibration as the primary levers for improvement suggests that several of these tasks may be approaching a plateau where architectural novelty yields diminishing returns, and further gains will require richer annotations, larger datasets, or multimodal grounding rather than model scaling alone.
  • The near-ceiling performance on VQA (98% accuracy) and athlete localization (97.67 mAP-LocSim) indicates these specific benchmarks may be approaching saturation, motivating the design of harder, more compositional evaluation protocols in future editions.
  • The strong showing of task-routed VLM pipelines—combining frontier general-purpose models with soccer-specific retrieval and lightweight perception tools—suggests a viable template for other specialized video understanding domains where end-to-end fine-tuning of large models is impractical.
  • The persistent difficulty with rare action classes (e.g., Tackle) and occluded players across multiple tasks points to a shared bottleneck: maintaining player identity and visual evidence through occlusions, which may require new tracking or temporal reasoning mechanisms rather than improved single-frame perception.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. This paper reports the results of the SoccerNet 2026 Challenges, the sixth annual edition of the SoccerNet benchmarking effort. It covers five vision-based tasks—Ball Action Anticipation (BAA), Player-Centric Ball Action Spotting (PCBAS), Novel View Synthesis (NVS), Spiideo SoccerNet Synloc (SSS), and Visual Question Answering (VQA)—describing each task's dataset, evaluation protocol, leaderboard on held-out challenge data, and the leading submissions' methods. Across all five tasks, the winning entries improved over the provided baselines. The paper also summarizes recurring methodological themes (higher input resolution, ensembling, domain-specific geometric/tactical features) and identifies where performance remains limited.

Significance. The paper serves as a standard challenge-results reference for the sports video understanding community, continuing a well-established series. Its strengths include transparent disclosure that only teams with reviewed technical reports are included in the leaderboards (§1.2), clear evaluation protocol definitions for each task, and a useful cross-task synthesis of methodological trends in §7. The breadth of participation (427 teams, 1,129 entries) and the inclusion of five diverse tasks make this a valuable community resource. The winning method summaries are sufficiently detailed to be informative for practitioners.

major comments (3)
  1. §2.1 and Table 1 (BAA): The challenge test set consists of only 2 matches. The gap between 1st place (24.08) and 2nd place (21.36) is 2.72 mAP points, and the gap between 2nd and 3rd is 0.22 points. With n=2 matches, match-level variance in action distributions could plausibly swing rankings, particularly for the 2nd-vs-3rd distinction. The paper does not acknowledge this limitation or discuss the reliability of the BAA leaderboard. A brief note on the small test-set size and its implications for ranking reliability would strengthen the paper's central claim that the leaderboards document the current state of each task.
  2. §6.2 and Table 5 (VQA): The challenge split contains 500 multiple-choice questions. The gap between 1st (98.0%) and 2nd (96.0%) is 10 questions, and between 2nd and 3rd is 5 questions. Approximate 95% binomial confidence intervals at n=500 for scores near 96–98% are roughly ±1.7%, meaning ranks 1–3 are not statistically distinguishable. The paper should acknowledge this and caution against over-interpreting small ranking differences, at least for this task.
  3. §6.4 (VQA winner) vs. Table 5: The winner's summary in the supplementary (§8.5, VQA-1) reports 97.6% accuracy, while Table 5 lists 98.0% for the same team (vitomeme). Similarly, the NVS winner summary (§4.4/§8.3, NVS-1) reports LPIPS of 0.366, while Table 3 lists 0.388. These discrepancies should be reconciled or explained.
minor comments (4)
  1. Table 5: The rank column jumps from 4 to 11 to 12 to 16 to 34. While the paper explains that only teams with technical reports are included, adding a footnote to the table itself would make this clearer at a glance.
  2. §5.2: The LocSim formula is rendered as 'e ln 0.05 d2 τ 2', which appears to be a formatting issue. The formula should be typeset clearly, likely as exp(ln(0.05) · d²/τ²) or equivalent.
  3. §4.2: The sentence beginning 'Regarding PSNR, it may favor Gaussian primitives...' is somewhat informal and could be tightened for clarity.
  4. The author list is extremely long (challenge participants). While this is standard for challenge papers, confirming that the metadata (affiliations, equal contribution markers) is correct for all listed authors would be advisable for the camera-ready.

Circularity Check

0 steps flagged

No circularity found. This is a benchmarking report whose central claim is verified against held-out data, not derived from fitted parameters or self-cited premises.

full rationale

The paper is a challenge results report for SoccerNet 2026, covering five tasks. Its central claim—that leading submissions improved over provided baselines—is directly verified by leaderboard scores on held-out challenge splits (e.g., BAA: 24.08 vs 16.76 baseline; PCBAS: 58.94 vs 46.41; NVS: 29.89 vs 26.74; SSS: 97.67 vs 77.30; VQA: 98.0% vs 25.0% random). No step in the paper's chain involves deriving a prediction from a fitted parameter and then presenting it as an independent result, defining a quantity in terms of what it claims to predict, or invoking a self-cited uniqueness theorem to force a conclusion. Self-citations exist for task and dataset definitions (e.g., [22] for BAA, [73] for PCBAS, [6] for SSS, [77] for VQA), and some authors overlap with prior SoccerNet publications. However, these citations define the benchmark setup (datasets, metrics, protocols), not the results. The results are computed externally by an evaluation server on private test data, making them independently falsifiable. This is standard for challenge papers and does not constitute circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

This is a challenge results paper, not a theoretical derivation. There are no free parameters fitted by the paper authors, no new axioms introduced, and no invented entities. The benchmarks and metrics are defined by prior work and standard practice.

axioms (2)
  • standard math The evaluation metrics (mAP, F1, PSNR, SSIM, LPIPS, accuracy) are valid measures of task performance.
    Standard metrics used across computer vision benchmarks.
  • domain assumption The held-out challenge test sets are representative of the broader problem domain.
    The paper assumes that performance on the challenge split reflects generalizable method capability, though some splits are small (e.g., 2 matches for BAA, 500 questions for VQA).

pith-pipeline@v1.1.0-glm · 37253 in / 1405 out tokens · 258479 ms · 2026-07-09T14:08:39.173301+00:00 · methodology

0 comments
read the original abstract

The SoccerNet 2026 Challenges constitute the sixth annual edition of the SoccerNet open benchmarking effort, dedicated to advancing computer vision research in sports video understanding. This year's challenges span five vision-based tasks: (1) Ball Action Anticipation, predicting the timing and class of ball-related actions within a short future window from a preceding observation window; (2) Player-Centric Ball Action Spotting, temporally localizing and classifying ball-related actions while assigning each action to the acting player through team affiliation and jersey number; (3) Novel View Synthesis, rendering images from unobserved camera poses in multi-view football scenes; (4) Spiideo SoccerNet Synloc, localizing athletes in real-world pitch coordinates from a single calibrated static-camera image; and (5) Visual Question Answering, answering multiple-choice questions about football broadcasts across text, image, and video inputs. For each task, participants were provided with annotated data, a unified evaluation protocol, and a public baseline. This edition saw broad participation, with 427 teams submitting 1,129 entries across the five tasks and 28 teams contributing reviewed technical reports. This paper describes each task and its evaluation protocol, presents the challenge leaderboards, and summarizes the leading submissions, with the aim of documenting the current state of each task as measured on held-out challenge data.

Figures

Figures reproduced from arXiv: 2607.07320 by Albert Clap\'es, Anthony Cioppa, Antoine Houet, Artem Konshin, Artur Xarles, Atom Scott, Bernard Ghanem, Bogdan Stanciulescu, C\'edric Hons, Christian Orduz, Di Yang, Fabian Perez, Faisal Altawijri, Falguni Ghosh, Fang Liu, H{\aa}kan Ard\"o, Haoxuan Ma, Hoang-Phuc Nguyen, Hoover Rueda-Chac\'on, Ho-Young Jung, Ikuma Uchida, Ismail Mathkour, Jakub Komosa, Jan Held, J\'er\'emie Ochin, Jiali Wen, Jiangtao Wang, Jianling Chu, Jiayuan Rao, Jongmin Lee, Juan Vanegas, Julian Ziegler, Ju-Seong Do, Karen Sanchez, Konstantin Mitin, Kwanyong Park, Lechao Cheng, Lingling Li, Litao Li, Marc Van Droogenbroeck, Mathieu Delvaux, Matvey Isupov, Mikael Nilsson, Mikhail Moiseev, Minh-Triet Tran, Minjae Kim, Minori Sugimura, Mirco Fuchs, Mohamad Dalal, Mohamed Atef, Oleg Durygin, Olivier Barnich, Omar Fetouh, Parthsarthi Rawat, Phuong-Linh Huynh-Ha, Pierre Miralles, Puhua Chen, Renaud Vandeghen, Rio Watanabe, Ruifeng Wang, Semen Budennyy, SeongHeon Kang, Sergio Escalera, Shengeng Tang, Shun Makino, Silvio Giancola, Siyuan Jiang, Sotiris Manitsaris, SuHyun Rim, Takumi Nagaya, Thanh-Khoi Nguyen, Thanh-Nhan Vo, Thomas B. Moeslund, Tom Michel, Trong-Thuan Nguyen, Trung-Hoang Le, Vadim Linkov, Vasiliy Chelpanov, Weidi Xie, Weixuan Huang, Wenbo Zhu, Wonjun Heo, Wonyong Jo, Xiaogang Wang, Xingyu Zhu, Xinyu Ye, Xu Yang, Yanfeng Wang, Yangguang Ji, Yaxiong Wang, Yibo Yu, Yixin Chen, Yixi Zhou, Yongliang Wu, Youngseon Kim, Youssef Ghallab, Yufeng Hu, Yuki Nakamura, Yuyang Sun, Yu Zhang, Zhenxiang Jiang, Zhenyu Zhao, Zhun Zhong, Zhuo Yang, Zihan Zhai.

Figure 1
Figure 1. Figure 1: Overview of the challenges. In 2026, the SoccerNet challenges encompass five vision tasks: (1) Ball Action Anticipation (BAA), focusing on predicting the timing and class of ball-related actions occurring within a five-second window from a preced￾ing 30-second observation window, (2) Player-Centric Ball Action Spotting (PCBAS), focusing on temporally localizing and classifying ball-related actions while as… view at source ↗

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

115 extracted references · 115 canonical work pages · 7 internal anchors

  1. [1]

    In: IEEE Int

    Akyon, F.C., Onur Altinuc, S., Temizel, A.: Slicing aided hyper infer- ence and fine-tuning for small object detection. In: IEEE Int. Conf. Im- age Process. (ICIP). pp. 966–970. IEEE, Bordeaux, France (Oct 2022). https://doi.org/10.1109/icip46576.2022.989799012, 35

    work page doi:10.1109/icip46576.2022.989799012 2022

  2. [2]

    Andrews, P., Nordberg, O.E., Zubicueta Portales, S., Borch, N., Guribye, F., Fujita, K., Fjeld, M.: AiCommentator: A multimodal conversational agent for embedded visualization in football viewing. In: Int. Conf. Intell. User Interfaces. pp. 14–34. ACM, Greenville, SC, USA (Mar 2024).https: //doi.org/10.1145/3640543.36451973

    work page doi:10.1145/3640543.36451973 2024

  3. [3]

    Anthropic:ClaudeSonnet.https://www.anthropic.com/claude/sonnet (2025) 15, 38

    work page 2025

  4. [4]

    In: IEEE Int

    Antol, S., Agrawal, A., Lu, J., Mitchell, M., Batra, D., Zitnick, C.L., Parikh, D.: VQA: Visual question answering. In: IEEE Int. Conf. Com- put. Vis. (ICCV). pp. 2425–2433. IEEE, Santiago, Chile (Dec 2015). https://doi.org/10.1109/iccv.2015.27913

    work page doi:10.1109/iccv.2015.27913 2015

  5. [5]

    In: IEEE/CVF Conf

    Arbués Sangüesa, A., Martín, A., Fernández, J., Ballester, C., Haro, G.: Using player’s body-orientation to model pass feasibility in soccer. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 3875–3884. IEEE, Seattle, WA, USA (Jun 2020).https://doi.org/10. 1109/cvprw50498.2020.004514

    work page arXiv 2020

  6. [6]

    Ardö, H., Nilsson, M., Cioppa, A., Magera, F., Giancola, S., Liu, H., Ghanem, B., Van Droogenbroeck, M.: Spiideo SoccerNet SynLoc: Sin- gle frame world coordinate athlete detection and localization with syn- thetic data. In: Int. Jt. Conf. Comput. Vis. Imaging Comput. Graph. The- ory Appl. vol. 2, pp. 278–285. SCITEPRESS - Science and Technology Publicati...

    work page 2025

  7. [7]

    Qwen3-VL Technical Report

    Bai, S., Cai, Y., Chen, R., Chen, K., Chen, X., Cheng, Z., Deng, L., Ding, W., Gao, C., Ge, C., Ge, W., Guo, Z., Huang, Q., Huang, J., Huang, F., Hui, B., Jiang, S., Li, Z., Li, M., Li, M., Li, K., Lin, Z., Lin, J., Liu, X., Liu, J., Liu, C., Liu, Y., Liu, D., Liu, S., Lu, D., Luo, R., Lv, C., Men, R., Meng, L., Ren, X., Ren, X., Song, S., Sun, Y., Tang, ...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2511.2163115 2025

  8. [8]

    Balaji, B., Bright, J., Prakash, H., Chen, Y., Clausi, D.A., Zelek, J.: Jer- sey number recognition using keyframe identification from low-resolution broadcast videos. In: Int. ACM Work. Multimedia Content Anal. Sports (MMSports). pp. 123–130. ACM, Ottawa, Ontario, Can. (Oct 2023). https://doi.org/10.1145/3606038.36161622 18 A. Cioppa et al

    work page doi:10.1145/3606038.36161622 2023

  9. [9]

    In: IEEE/CVF Conf

    Barron, J.T., Mildenhall, B., Verbin, D., Srinivasan, P.P., Hedman, P.: Mip-NeRF 360: Unbounded anti-aliased neural radiance fields. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR). pp. 5460–

    work page

  10. [10]

    In: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    IEEE, New Orleans, LA, USA (Jun 2022).https://doi.org/10. 1109/cvpr52688.2022.005399

    work page arXiv 2022

  11. [11]

    In: IEEE/CVF Conf

    Bou, X., Correger, N., Cloots, A., Gavage, C., Giancola, S., Schwartz, C., Delvaux, F., Cloots, R., Van Droogenbroeck, M., Cioppa, A.: Towards athlete fatigue assessment from association football videos. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 1–13. IEEE, Denver, CO, USA (Jun 2026) 4

    work page 2026

  12. [12]

    Are NeRFs Ready for Autonomous Driving? Towards Closing the Real-to-Simulation Gap

    Cabado, B., Cioppa, A., Giancola, S., Villa, A., Guijarro-Berdiñas, B., Padrón, E.J., Ghanem, B., Van Droogenbroeck, M.: Beyond the Pre- mier: Assessing action spotting transfer capability across diverse domains. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). vol. 27, pp. 3386–3398. IEEE, Seattle, WA, USA (Jun 2024).https: //doi.org/10.1...

    work page doi:10.1109/cvprw63382.2024.003432 2024

  13. [13]

    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 2026

  14. [14]

    Cheng, K., Long, X., Yang, K., Yao, Y., Yin, W., Ma, Y., Wang, W., Chen, X.: GaussianPro: 3D Gaussian splatting with progressive propagation. In: Int. Conf. Mach. Learn. (ICML). Proc. Mach. Learn. Res., vol. 235, pp. 123–8140. ML Res. Press (2024) 11, 34

    work page 2024

  15. [15]

    Cioppa, A., Deliège, A., Giancola, S., Ghanem, B., Van Droogenbroeck, M.: Scaling up SoccerNet with multi-view spatial localization and re- identification. Sci. Data9(1), 1–9 (Jun 2022).https://doi.org/10. 1038/s41597-022-01469-13, 4, 14

    work page 2022

  16. [16]

    In: Conference on Computer Vision and Pattern Recognition

    Cioppa, A., Deliège, A., Giancola, S., Ghanem, B., Van Droogenbroeck, M., Gade, R., Moeslund, T.B.: A context-aware loss function for action spotting in soccer videos. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR). pp. 13123–13133. IEEE, Seattle, WA, USA (Jun 2020). https://doi.org/10.1109/cvpr42600.2020.013142

    work page doi:10.1109/cvpr42600.2020.013142 2020

  17. [17]

    In: IEEE Int

    Cioppa, A., Deliège, A., Giancola, S., Magera, F., Barnich, O., Ghanem, B., Van Droogenbroeck, M.: Camera calibration and player localization in SoccerNet-v2 and investigation of their representations for action spotting. In: IEEE Int. Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW), CVsports. pp. 4532–4541. Nashville, TN, USA (Jun 2021).https://doi. o...

    work page doi:10.1109/cvprw53098.2021.005112 2021

  18. [18]

    In: IEEE Int

    Cioppa, A., Deliège, A., Istasse, M., De Vleeschouwer, C., Van Droogen- broeck, M.: ARTHuS: Adaptive real-time human segmentation in sports through online distillation. In: IEEE Int. Conf. Comput. Vis. Pattern SN2026 19 Recognit. Work. (CVPRW), CVsports. pp. 2505–2514. IEEE, Long Beach, CA, USA (Jun 2019).https://doi.org/10.1109/cvprw.2019.003062

    work page doi:10.1109/cvprw.2019.003062 2019

  19. [19]

    In: IEEE Int

    Cioppa, A., Deliège, A., Van Droogenbroeck, M.: A bottom-up approach based on semantics for the interpretation of the main camera stream in soccer games. In: IEEE Int. Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW), CVsports. pp. 1846–1855. Salt Lake City, UT, USA (Jun 2018). https://doi.org/10.1109/CVPRW.2018.002292

    work page doi:10.1109/cvprw.2018.002292 2018

  20. [20]

    In: IEEE Int

    Cioppa, A., Giancola, S., Deliège, A., Kang, L., Zhou, X., Cheng, Z., Ghanem, B., Van Droogenbroeck, M.: SoccerNet-tracking: Multiple ob- ject tracking dataset and benchmark in soccer videos. In: IEEE Int. Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW), CVsports. pp. 3490–

    work page

  21. [21]

    Desi: Deepfake source identifier for social media,

    IEEE, New Orleans, LA, USA (Jun 2022).https://doi.org/10. 1109/cvprw56347.2022.003934

    work page arXiv 2022

  22. [22]

    SoccerNet 2024 Challenges Results

    Cioppa, A., Giancola, S., Somers, V., Joos, V., Magera, F., Held, J., Ghasemzadeh, S.A., Zhou, X., Seweryn, K., Kowalczyk, M., Mróz, Z., Łukasik, S., Hałoń, M., Mkhallati, H., Deliège, A., Hinojosa, C., Sanchez, K., Mansourian, A.M., Miralles, P., Barnich, O., De Vleeschouwer, C., Alahi, A., Ghanem, B., Van Droogenbroeck, M., Gorski, A., Clapés, A., Boiar...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2409.105873 2024

  23. [23]

    Cioppa et al

    Cioppa, A., Giancola, S., Somers, V., Magera, F., Zhou, X., Mkhallati, H., Deliège, A., Held, J., Hinojosa, C., Mansourian, A.M., Miralles, P., Barnich, O., De Vleeschouwer, C., Alahi, A., Ghanem, B., Van Droogen- broeck, M., Kamal, A., Maglo, A., Clapés, A., Abdelaziz, A., Xarles, A., Orcesi, A., Scott, A., Liu, B., Lim, B., Chen, C., Deuser, F., Yan, F....

    work page doi:10.1007/s12283-024-00466-43 2023

  24. [25]

    In: IEEE Int

    Deliège, A., Cioppa, A., Giancola, S., Seikavandi, M.J., Dueholm, J.V., Nasrollahi, K., Ghanem, B., Moeslund, T.B., Van Droogenbroeck, M.: SoccerNet-v2: A dataset and benchmarks for holistic understanding of broadcast soccer videos. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 4503–4514. IEEE, Nashville, TN, USA (Jun 2021).https://...

    work page doi:10.1109/cvprw53098.2021.005084 2021

  25. [26]

    IEEE Trans

    Deng, J., Guo, J., Yang, J., Xue, N., Kotsia, I., Zafeiriou, S.: ArcFace: Ad- ditive angular margin loss for deep face recognition. IEEE Trans. Pattern Anal. Mach. Intell.44(10), 5962–5979 (Oct 2022).https://doi.org/10. 1109/tpami.2021.308770915

    work page arXiv 2022

  26. [27]

    Falaleev,N.S.,Chen,R.:Enhancingsoccercameracalibrationthroughkey- point exploitation. In: Int. ACM Work. Multimedia Content Anal. Sports (MMSports). vol. 6, pp. 65–73. ACM, Melbourne, Victoria, Aust. (Oct 2024).https://doi.org/10.1145/3689061.36890743

    work page doi:10.1145/3689061.36890743 2024

  27. [28]

    Sports Eng.28(2), 1–10 (Jul 2025).https://doi

    Fang, J., Yeung, C., Fujii, K.: Foul prediction with estimated poses from soccer broadcast video. Sports Eng.28(2), 1–10 (Jul 2025).https://doi. org/10.1007/s12283-025-00515-62

    work page doi:10.1007/s12283-025-00515-62 2025

  28. [29]

    In: Conference on Computer Vision and Pattern Recognition

    Feichtenhofer, C.: X3D: Expanding architectures for efficient video recog- nition. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR). pp. 200–210. IEEE, New Orleans, LA, USA (Jun 2020).https://doi.org/ 10.1109/cvpr42600.2020.0002833

    work page doi:10.1109/cvpr42600.2020.0002833 2020

  29. [30]

    Gade, R., Merler, M., Thomas, G., Moeslund, T.B.: The (Computer) Vision of Sports: Recent Trends in Research and Commercial Systems for Sport Analytics, chap. 14, pp. 1–16. Chapman and Hall/CRC, New York City, NY, USA (2024).https://doi.org/10.1201/9781003328957, https://doi.org/10.1201/97810033289572, 3

    work page doi:10.1201/9781003328957 2024

  30. [31]

    In: Work

    Gautam, S., Midoglu, C., Shafiee Sabet, S., Kshatri, D.B., Halvorsen, P.: Soccer game summarization using audio commentary, metadata, and cap- tions. In: Work. User-centric Narrat. Summ. Long Videos (narsum). pp. 13–22. ACM, Lisboa Portugal (Oct 2022).https://doi.org/10.1145/ 3552463.35570194

    work page arXiv 2022

  31. [32]

    Gautam, S., Sarkhoosh, M.H., Held, J., Midoglu, C., Cioppa, A., Giancola, S., Thambawita, V., Riegler, M.A., Halvorsen, P., Shah, M.: SoccerNet- Echoes: A soccer game audio commentary dataset. Int. Symp. Multime- dia (ISM) pp. 71–78 (Dec 2024).https://doi.org/10.1109/ism63611. 2024.000163, 4 SN2026 21

    work page doi:10.1109/ism63611 2024

  32. [34]

    In: IEEE/CVF Conf

    Giancola, S., Amine, M., Dghaily, T., Ghanem, B.: SoccerNet: A scalable dataset for action spotting in soccer videos. In: IEEE/CVF Conf. Com- put. Vis. Pattern Recognit. Work. (CVPRW). pp. 1792–179210. IEEE, Salt Lake City, UT, USA (Jun 2018).https://doi.org/10.1109/cvprw. 2018.002233

    work page doi:10.1109/cvprw 2018

  33. [35]

    Giancola, S., Cioppa, A., Deliège, A., Magera, F., Somers, V., Kang, L., Zhou, X., Barnich, O., De Vleeschouwer, C., Alahi, A., Ghanem, B., Van Droogenbroeck, M., Darwish, A., Maglo, A., Clapés, A., Luyts, A., Boiarov, A., Xarles, A., Orcesi, A., Shah, A., Fan, B., Comandur, B., Chen, C., Zhang, C., Zhao, C., Lin, C., Chan, C.Y., Hui, C.C., Li, D., Yang, ...

    work page doi:10.1145/3552437.35585453 2022

  34. [36]

    Topology-aware focal loss for 3d image segmentation,

    Giancola, S., Cioppa, A., Georgieva, J., Billingham, J., Serner, A., Peek, K., Ghanem, B., Van Droogenbroeck, M.: Towards active learning for ac- tion spotting in association football videos. In: IEEE/CVF Conf. Comput. Vis.PatternRecognit.Work.(CVPRW).pp.5098–5108.IEEE,Vancouver, Can. (Jun 2023).https://doi.org/10.1109/cvprw59228.2023.005382

    work page doi:10.1109/cvprw59228.2023.005382 2023

  35. [37]

    Com- put

    Giancola, S., Cioppa, A., Ghanem, B., Van Droogenbroeck, M.: Deep learning for action spotting in association football videos, Ser. Com- put. Vis., vol. 9, chap. 2.5, pp. 427–459. World Sci. (Jul 2025).https: //doi.org/10.1142/9789819807154_0018,https://doi.org/10.1142/ 9789819807154_00182

    work page doi:10.1142/9789819807154_0018 2025

  36. [38]

    SoccerNet 2025 Challenges Results

    Giancola, S., Cioppa, A., Gutiérrez-Pérez, M., Held, J., Hinojosa, C., Joos, V., Leduc, A., Magera, F., Sanchez, K., Somers, V., Xarles, A., Agudo, A., Alahi, A., Barnich, O., Clapés, A., De Vleeschouwer, C., Es- calera, S., Ghanem, B., Moeslund, T.B., Van Droogenbroeck, M., Abe, T., Alotaibi, S., Altawijri, F., Araujo, S., Bai, X., Bi, X., Cao, J., Chao,...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2508.191823 2025

  37. [39]

    In: Conf

    Gu, A., Dao, T.: Mamba: Linear-time sequence modeling with selective state spaces. In: Conf. Lang. Model. pp. 1–32. Philadelphia, PA, USA (Oct 2024) 32

    work page 2024

  38. [40]

    Are NeRFs Ready for Autonomous Driving? Towards Closing the Real-to-Simulation Gap

    Gutiérrez-Pérez, M., Agudo, A.: No bells, just whistles: Sports field regis- tration by leveraging geometric properties. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 3325–3334. IEEE, Seattle, WA, USA (Jun 2024).https://doi.org/10.1109/cvprw63382.2024. 003373

    work page doi:10.1109/cvprw63382.2024 2024

  39. [41]

    In: IEEE/CVF Conf

    Gutiérrez-Pérez, M., Agudo, A.: SoccerNet-v3D: Leveraging sports broad- cast replays for 3D scene understanding. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 5968–5977. IEEE, Nashville, TN, USA (Jun 2025).https://doi.org/10.1109/cvprw67362.2025. 005954

    work page doi:10.1109/cvprw67362.2025 2025

  40. [42]

    Gutiérrez-Pérez, M., Agudo, A.: PnLCalib: Sports field registration via points and lines optimization. Comput. Vis. Image Underst.267, 104712 (Apr 2026).https://doi.org/10.1016/j.cviu.2026.1047123

    work page doi:10.1016/j.cviu.2026.1047123 2026

  41. [43]

    IEEE Trans

    Hahne, C., Aggoun, A.: PlenoptiCam v1.0: A light-field imaging frame- work. IEEE Trans. Image Process.30, 6757–6771 (2021).https://doi. org/10.1109/tip.2021.309567135

    work page doi:10.1109/tip.2021.309567135 2021

  42. [44]

    In: MathSport Conference

    Held, J., Cioppa, A., Giancola, S., Almahmoud, E., Collins, K.M., Bhatt, U., Ghanem, B., Van Droogenbroeck, M.: Enhancing football refereeing with AI: VARS and X-VARS for assisted decision-making. In: MathSport Conference. Luxembourg (Jun 2025) 2

    work page 2025

  43. [45]

    In: Scand

    Held, J., Cioppa, A., Giancola, S., Hamdi, A., Devue, C., Ghanem, B., Van Droogenbroeck, M.: Towards an AI-powered video assistant referee system (VARS) for association football. In: Scand. Conf. Image Anal. (SCIA). Lect. Notes Comput. Sci., vol. 15725, pp. 295–309. Springer Nat. Switz. (2025).https://doi.org/10.1007/978-3-031-95911-0_21

    work page doi:10.1007/978-3-031-95911-0_21 2025

  44. [46]

    Topology-aware focal loss for 3d image segmentation,

    Held, J., Cioppa, A., Giancola, S., Hamdi, A., Ghanem, B., Van Droogen- broeck, M.: VARS: Video assistant referee system for automated soccer decision making from multiple views. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 5086–5097. IEEE, Vancouver, Can. (Jun 2023).https://doi.org/10.1109/cvprw59228.2023.00537 4

    work page doi:10.1109/cvprw59228.2023.00537 2023

  45. [47]

    Are NeRFs Ready for Autonomous Driving? Towards Closing the Real-to-Simulation Gap

    Held, J., Itani, H., Cioppa, A., Giancola, S., Ghanem, B., Van Droogen- broeck, M.: X-VARS: Introducing explainability in football refereeing with SN2026 23 multi-modal large language models. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 3267–3279. IEEE, Seattle, WA, USA (Jun 2024).https://doi.org/10.1109/cvprw63382.2024.00332 2, 4, 14

    work page doi:10.1109/cvprw63382.2024.00332 2024

  46. [48]

    In: IEEE/CVF Conf

    Held, J., Son, S., Vandeghen, R., Rebain, D., Gadelha, M., Zhou, Y., Cioppa, A., Lin, M.C., Van Droogenbroeck, M., Tagliasacchi, A.: Mesh- Splatting: Differentiable rendering with opaque meshes. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR). IEEE, Denver, CO, USA (Jun 2026) 9

    work page 2026

  47. [49]

    Held, J., Vandeghen, R., Deliège, A., Hamdi, Abdullah Rebain, D., Giancola, S., Cioppa, A., Vedaldi, A., Ghanem, B., Tagliasacchi, A., Van Droogenbroeck, M.: Triangle splatting for real-time radiance field ren- dering. In: Int. Conf. 3D Vis. (3DV). pp. 1–10. Vancouver, Can. (Mar 2026) 9, 10, 11, 34, 35

    work page 2026

  48. [50]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference (2025)

    Held, J., Vandeghen, R., Hamdi, A., Deliège, A., Cioppa, A., Giancola, S., Vedaldi, A., Ghanem, B., Van Droogenbroeck, M.: 3D convex splat- ting: Radiance field rendering with 3D smooth convexes. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR). pp. 21360–21369. IEEE, Nashville, TN, USA (Jun 2025).https://doi.org/10.1109/CVPR52734. 2025.019909

    work page doi:10.1109/cvpr52734 2025

  49. [51]

    Desi: Deepfake source identifier for social media,

    Honda, Y., Kawakami, R., Yoshihashi, R., Kato, K., Naemura, T.: Pass receiver prediction in soccer using video and players’ trajectories. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 3502–3511. IEEE, New Orleans, LA, USA (Jun 2022).https://doi.org/ 10.1109/cvprw56347.2022.003944

    work page doi:10.1109/cvprw56347.2022.003944 2022

  50. [52]

    Hong, J., Zhang, H., Gharbi, M., Fisher, M., Fatahalian, K.: Spotting temporally precise, fine-grained events in video. In: Eur. Conf. Comput. Vis. (ECCV). Lect. Notes Comput. Sci., vol. 13695, pp. 33–51. Springer Nat. Switz., Tel Aviv, Israël (2022).https://doi.org/10.1007/978-3- 031-19833-5_32

    work page doi:10.1007/978-3- 2022

  51. [53]

    Hu, E.J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., Chen, W.: LoRA: Low-rank adaptation of large language models. In: Int. Conf. Learn. Represent. (ICLR). pp. 1–13. Virtual conference (Sept 2022) 15

    work page 2022

  52. [54]

    In: ACM SIGGRAPH Conf

    Huang, B., Yu, Z., Chen, A., Geiger, A., Gao, S.: 2D Gaussian splatting for geometrically accurate radiance fields. In: ACM SIGGRAPH Conf. Pap. vol. 35, pp. 1–11. ACM, Denver, CO, USA (Jul 2024).https://doi.org/ 10.1145/3641519.365742811

    work page doi:10.1145/3641519.365742811 2024

  53. [56]

    Jiang, T., Billingham, J., Müksch, S., Zarate, J., Evans, N., Oswald, M.R., Polleyfeys, M., Hilliges, O., Kaufmann, M., Song, J.: WorldPose: A world cup dataset for global 3D human pose estimation. In: Eur. Conf. Comput. Vis. (ECCV). Lect. Notes Comput. Sci., vol. 15077, pp. 343–362. Springer 24 A. Cioppa et al. Nat. Switz. (Dec 2024).https://doi.org/10.1...

    work page doi:10.1007/978-3-031-72655- 2024

  54. [57]

    In:Int.ACM Work

    Jiang, Y., Cui, K., Chen, L., Wang, C., Xu, C.: SoccerDB: A large-scale databasefor comprehensive videounderstanding. In:Int.ACM Work. Mul- timedia Content Anal. Sports (MMSports). pp. 1–8. ACM, Seattle, WA, USA (Oct 2020).https://doi.org/10.1145/3422844.34230514

    work page doi:10.1145/3422844.34230514 2020

  55. [58]

    Pixels or Positions? Benchmarking Modalities in Group Activity Recognition

    Karki, D., Ramazanova, M., Cioppa, A., Giancola, S., Ghanem, B.: Pix- els or positions? benchmarking modalities in group activity recognition. arXivabs/2511.12606(2025).https://doi.org/10.48550/arXiv. 2511.126062, 4

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv 2025

  56. [59]

    ACM Trans

    Kerbl, B., Kopanas, G., Leimkuehler, T., Drettakis, G.: 3D Gaussian splat- ting for real-time radiance field rendering. ACM Trans. Graph.42(4), 1–14 (Jul 2023).https://doi.org/10.1145/35924339, 10, 11, 34

    work page doi:10.1145/35924339 2023

  57. [60]

    Kheradmand, S., Rebain, D., Sharma, G., Sun, W., Tseng, J., Isack, H., Kar, A., Tagliasacchi, A., Yi, K.M.: 3D Gaussian splatting as Markov chain Monte Carlo. In: Adv. Neural Inf. Process. Syst. (NeurIPS). vol. 37, pp. 80965–80986. Curran Assoc. Inc., Vancouver, Can. (Dec 2024) 9

    work page 2024

  58. [61]

    ACM Trans

    Knapitsch,A.,Park,J.,Zhou,Q.Y.,Koltun,V.:Tanksandtemples:bench- marking large-scale scene reconstruction. ACM Trans. Graph.36(4), 1–13 (Jul 2017).https://doi.org/10.1145/3072959.30735999

    work page doi:10.1145/3072959.30735999 2017

  59. [62]

    Are NeRFs Ready for Autonomous Driving? Towards Closing the Real-to-Simulation Gap

    Leduc, A., Cioppa, A., Giancola, S., Ghanem, B., Van Droogenbroeck, M.: SoccerNet-Depth: a scalable dataset for monocular depth estimation in sports videos. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). vol. 12, pp. 3280–3282. IEEE, Seattle, WA, USA (Jun 2024).https://doi.org/10.1109/cvprw63382.2024.003334

    work page doi:10.1109/cvprw63382.2024.003334 2024

  60. [63]

    Lin, H., Chen, S., Liew, J., Chen, D.Y., Li, Z., Shi, G., Feng, J., Kang, B.: Depth Anything 3: Recovering the visual space from any views. In: Int. Conf. Learn. Represent. (ICLR). pp. 1–25. Rio De Janeiro, Braz. (Apr

    work page

  61. [64]

    10 João Carreira, Dilara Gokay, Michael King, Chuhan Zhang, Ignacio Rocco, Aravindh Mahendran, Thomas Albert Keck, et al

    Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., Guo, B.: Swin transformer: Hierarchical vision transformer using shifted windows. In: IEEE/CVF Int. Conf. Comput. Vis. (ICCV). pp. 9992–10002. IEEE, Mon- tréal, Can. (Oct 2021).https://doi.org/10.1109/iccv48922.2021. 0098633

    work page doi:10.1109/iccv48922.2021 2021

  62. [65]

    Ludwig, K.: Human pose estimation in images and videos for sports ana- lytics: 2D keypoint and 3D mesh estimation for challenging scenarios and extreme poses. Ph.D. thesis, Universität Augsburg, Germany (Jul 2025) 2

    work page 2025

  63. [66]

    In: IEEE/CVF Conf

    Magera, F., Hoyoux, T., Barnich, O., Van Droogenbroeck, M.: A universal protocol to benchmark camera calibration for sports. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 3335–3346. IEEE, Seattle, WA, USA (Jun 2024).https://doi.org/10.1109/cvprw63382. 2024.003383

    work page doi:10.1109/cvprw63382 2024

  64. [67]

    Magera,F.,Hoyoux,T.,Castin,M.,Barnich,O.,Cioppa,A.,VanDroogen- broeck, M.: Can geometry save central views for sports field registration? In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). SN2026 25 pp. 6060–6069. IEEE, Nashville, TN, USA (Jun 2025).https://doi.org/ 10.1109/cvprw67362.2025.006043

    work page doi:10.1109/cvprw67362.2025.006043 2025

  65. [68]

    Desi: Deepfake source identifier for social media,

    Maglo, A., Orcesi, A., Pham, Q.C.: Efficient tracking of team sport players with few game-specific annotations. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 3460–3470. IEEE, New Orleans, LA, USA (Jun 2022).https://doi.org/10.1109/cvprw56347.2022. 003902

    work page doi:10.1109/cvprw56347.2022 2022

  66. [69]

    Desi: Deepfake source identifier for social media,

    Maji, D., Nagori, S., Mathew, M., Poddar, D.: YOLO-pose: Enhancing YOLO for multi person pose estimation using object keypoint similar- ity loss. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 2636–2645. IEEE, New Orleans, LA, USA (Jun 2022). https://doi.org/10.1109/cvprw56347.2022.0029713

    work page doi:10.1109/cvprw56347.2022.0029713 2022

  67. [70]

    Mansourian, A.M., Somers, V., De Vleeschouwer, C., Kasaei, S.: Multi- task learning for joint re-identification, team affiliation, and role classifi- cation for sports visual tracking. In: Int. ACM Work. Multimedia Content Anal. Sports (MMSports). pp. 103–112. ACM, Ottawa, Ontario, Can. (Oct 2023).https://doi.org/10.1145/3606038.36161722

    work page doi:10.1145/3606038.36161722 2023

  68. [71]

    In: Proceedings of the Mile-High Video Conference

    Midoglu, C., Sabet, S.S., Sarkhoosh, M.H., Majidi, M., Gautam, S., Sol- berg, H.M., Kupka, T., Halvorsen, P.: AI-based sports highlight generation for social media. In: Proceedings of the Mile-High Video Conference. pp. 7–13. ACM, Denver, CO, USA (Feb 2024).https://doi.org/10.1145/ 3638036.36407994

    work page arXiv 2024

  69. [72]

    Topology-aware focal loss for 3d image segmentation,

    Mkhallati, H., Cioppa, A., Giancola, S., Ghanem, B., Van Droogenbroeck, M.: SoccerNet-caption: Dense video captioning for soccer broadcasts com- mentaries. In: IEEE/CVF Conf. Comput. Vis. Pattern Recognit. Work. (CVPRW). pp. 5074–5085. IEEE, Vancouver, Can. (Jun 2023).https: //doi.org/10.1109/cvprw59228.2023.005364, 14, 15

    work page doi:10.1109/cvprw59228.2023.005364 2023

  70. [74]

    Ochin, J., Chekroun, R., Stanciulescu, B., Manitsaris, S.: Beyond pixels: Leveraging the language of soccer to improve spatio-temporal action de- tection in broadcast videos. In: Adv. Concepts Intell. Vis. Syst. (ACIVS). Lect. Notes Comput. Sci., vol. 15656, pp. 552–563. Springer Nat. Switz. (2025).https://doi.org/10.1007/978-3-032-07343-3_447, 8, 33

    work page doi:10.1007/978-3-032-07343-3_447 2025

  71. [75]

    Ochin, J., Chekroun, R., Stanciulescu, B., Manitsaris, S.: FOOTPASS: A multi-modal multi-agent tactical context dataset for play-by-play action spotting in soccer broadcast videos. Comput. Vis. Image Underst.269, 1–13 (Jun 2026).https://doi.org/10.1016/j.cviu.2026.1047904, 7, 33

    work page doi:10.1016/j.cviu.2026.1047904 2026

  72. [76]

    Ochin, J., Devineau, G., Stanciulescu, B., Manitsaris, S.: Game state and spatio-temporal action detection in soccer using graph neural networks and 3D convolutional networks. In: Int. Conf. Pattern Recognit. Appl. Methods (ICPRAM). pp. 636–646. SCITEPRESS - Sci. Technol. Publ., Porto, Port. (2025).https://doi.org/10.5220/001316110000390533 26 A. Cioppa et al

    work page doi:10.5220/001316110000390533 2025

  73. [77]

    Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., Krueger, G., Sutskever, I.: Learning transferable visual models from natural language supervision. In: Int. Conf. Mach. Learn. (ICML). Proc. Mach. Learn. Res., vol. 139, pp. 8748–8763. ML Res. Press, Virtual Conf. (Jul 2021) 15

    work page 2021

  74. [78]

    GenTac: Generative Modeling and Forecasting of Soccer Tactics

    Rao, J., Gui, T., Wu, H., Wang, Y., Xie, W.: GenTac: Generative modeling and forecasting of soccer tactics. arXivabs/2604.11786(2026).https: //doi.org/10.48550/arXiv.2604.117864

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2604.117864 2026

  75. [79]

    In: ACM Int

    Rao, J., Li, Z., Wu, H., Zhang, Y., Wang, Y., Xie, W.: Multi-agent system for comprehensive soccer understanding. In: ACM Int. Conf. Multimedia (MM). pp. 3654–3663. ACM, Dublin, Irel. (Oct 2025).https://doi.org/ 10.1145/3746027.37551443, 13, 14, 15, 38

    work page doi:10.1145/3746027.37551443 2025

  76. [80]

    Rao, J., Wu, H., Jiang, H., Zhang, Y., Wang, Y., Xie, W.: Towards univer- salsoccervideounderstanding.In:IEEE/CVFConf.Comput.Vis.Pattern Recognit. (CVPR). pp. 8384–8394. IEEE, Nashville, TN, USA (Jun 2025). https://doi.org/10.1109/cvpr52734.2025.007854, 14, 15

    work page doi:10.1109/cvpr52734.2025.007854 2025

  77. [81]

    In: Proceedings of the Conference on Empirical Methods in Natural Language Processing

    Rao, J., Wu, H., Liu, C., Wang, Y., Xie, W.: MatchTime: Towards automatic soccer game commentary generation. In: Proceedings of the Conference on Empirical Methods in Natural Language Processing. pp. 1671–1685. Assoc. Comput. Linguistics, Miami, FL, USA (2024).https: //doi.org/10.18653/v1/2024.emnlp-main.993, 14

    work page doi:10.18653/v1/2024.emnlp-main.993 2024

  78. [82]

    Ravi, N., Gabeur, V., Hu, Y.T., Hu, R., Ryali, C., Ma, T., Khedr, H., Rädle, R., Rolland, C., Gustafson, L., Mintun, E., Pan, J., Alwala, K.V., Carion, N., Wu, C.Y., Girshick, R., Dollar, P., Feichtenhofer, C.: SAM 2: Segment anything in images and videos. In: Int. Conf. Learn. Represent. (ICLR). pp. 1–44. Singapore (Feb 2025) 15

    work page 2025

  79. [83]

    You only look once: Unified, real-time object detection

    Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: Unified, real-time object detection. In: IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR). pp. 779–788. Inst. Electr. Electron. Eng. (IEEE), Las Vegas, NV, USA (Jun 2016).https://doi.org/10.1109/cvpr.2016.91 15

    work page doi:10.1109/cvpr.2016.91 2016

  80. [84]

    Robinson, I., Robicheaux, P., Popov, M., Ramanan, D., Peri, P.: RF- DETR: Neural architecture search for real-time detection transformers. In: Int. Conf. Learn. Represent. (ICLR). pp. 1–14. Rio De Janeiro, Braz. (Apr 2026) 13

    work page 2026

Showing first 80 references.