SoccerNet-Tracking: Multiple Object Tracking Dataset and Benchmark in Soccer Videos
Not on the map. Why?
Abstract (arXiv)
Tracking objects in soccer videos is extremely important to gather both player and team statistics, whether it is to estimate the total distance run, the ball possession or the team formation. Video processing can help automating the extraction of those information, without the need of any invasive sensor, hence applicable to any team on any stadium. Yet, the availability of datasets to train learnable models and benchmarks to evaluate methods on a common testbed is very limited. In this work, we propose a novel dataset for multiple object tracking composed of 200 sequences of 30s each, representative of challenging soccer scenarios, and a complete 45-minutes half-time for long-term tracking. The dataset is fully annotated with bounding boxes and tracklet IDs, enabling the training of MOT baselines in the soccer domain and a full benchmarking of those methods on our segregated challenge sets. Our analysis shows that multiple player, referee and ball tracking in soccer videos is far from being solved, with several improvement required in case of fast motion or in scenarios of severe occlusion.
Key facts. Cited by 80 papers in the map (data through 14 Oct 2026). Builds on 31 of its 83 references that are in the map. First posted on arXiv 14 Apr 2022.
Builds on
31 of 83 references.
References · 31
- 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
- 2021YOLOX: Exceeding YOLO Series in 2021cited by 1,071
- 2016Simple Online and Realtime Trackingcited by 868
- 2021ByteTrack: Multi-Object Tracking by Associating Every Detection Boxcited by 823
- 2020HOTA: A Higher Order Metric for Evaluating Multi-Object Trackingcited by 552
- 2016MOT16: A Benchmark for Multi-Object Trackingcited by 532
- 2020FairMOT: On the Fairness of Detection and Re-Identification in Multiple Object Trackingcited by 490
- 2019Tracking without bells and whistlescited by 382
- 2020MOT20: A benchmark for multi object tracking in crowded scenescited by 357
- 2019Towards Real-Time Multi-Object Trackingcited by 311
- 2018CrowdHuman: A Benchmark for Detecting Human in a Crowdcited by 284
- 2016Joint Detection and Identification Feature Learning for Person Searchcited by 190
19 more
- 2017CityPersons: A Diverse Dataset for Pedestrian Detectioncited by 172
- 2014Multiple Object Tracking: A Literature Reviewcited by 166
- 2016Person Re-identification in the Wildcited by 157
- 2018SoccerNet: A Scalable Dataset for Action Spotting in Soccer Videoscited by 150
- 2020TAO: A Large-Scale Benchmark for Tracking Any Objectcited by 147
- 2015MOTChallenge 2015: Towards a Benchmark for Multi-Target Trackingcited by 133
- 2020SoccerNet-v2: A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videoscited by 131
- 2015UA-DETRAC: A New Benchmark and Protocol for Multi-Object Detection and Trackingcited by 98
- 2018Deep Affinity Network for Multiple Object Trackingcited by 82
- 2019A Context-Aware Loss Function for Action Spotting in Soccer Videoscited by 56
- 2021Camera Calibration and Player Localization in SoccerNet-v2 and Investigation of their Representations for Action Spottingcited by 43
- 2021Tracking Pedestrian Heads in Dense Crowdcited by 42
- 20203D-ZeF: A 3D Zebrafish Tracking Benchmark Datasetcited by 24
- 2020Using Player's Body-Orientation to Model Pass Feasibility in Soccercited by 20
- 2020Group Activity Detection from Trajectory and Video Data in Soccercited by 17
- 2021Contrastive Learning for Sports Video: Unsupervised Player Classificationcited by 17
- 2020Multimodal and multiview distillation for real-time player detection on a football fieldcited by 12
- 2020Self-Supervised Small Soccer Player Detection and Trackingcited by 12
- 2021Detecting and Matching Related Objects with One Proposal Multiple Predictionscited by 4
Built on by
Citing papers per year: 2022: 4; 2023: 16; 2024: 23; 2025: 24; 2026: 13.
80 papers.
Sort:
By area:
- 2023SportsMOT: A Large Multi-Object Tracking Dataset in Multiple Sports Scenescited by 131
- 2022Hard to Track Objects with Irregular Motions and Similar Appearances? Make It Easier by Buffering the Matching Spacecited by 66
- 2023Single-Model and Any-Modality for Video Object Trackingcited by 65
- 2023SoccerNet-Caption: Dense Video Captioning for Soccer Broadcasts Commentariescited by 57
- 2023VARS: Video Assistant Referee System for Automated Soccer Decision Making from Multiple Viewscited by 38
- 2022DeepSportradar-v1: Computer Vision Dataset for Sports Understanding with High Quality Annotationscited by 37
- 2024MatchTime: Towards Automatic Soccer Game Commentary Generationcited by 37
- 2023Iterative Scale-Up ExpansionIoU and Deep Features Association for Multi-Object Tracking in Sportscited by 29
- 2022SoccerNet 2022 Challenges Resultscited by 28
- 2024X-VARS: Introducing Explainability in Football Refereeing with Multi-Modal Large Language Modelcited by 22
- 2024Towards Universal Soccer Video Understandingcited by 21
- 2023SoccerNet 2023 Challenges Resultscited by 20
38 more
- 2023Towards Active Learning for Action Spotting in Association Football Videoscited by 14
- 2024A Universal Protocol to Benchmark Camera Calibration for Sportscited by 14
- 2024Multi-task Learning for Joint Re-identification, Team Affiliation, and Role Classification for Sports Visual Trackingcited by 13
- 2024Visible-Thermal Multiple Object Tracking: Large-scale Video Dataset and Progressive Fusion Approachcited by 13
- 2025TimeSoccer: An End-to-End Multimodal Large Language Model for Soccer Commentary Generationcited by 13
- 2024SoccerNet-Echoes: A Soccer Game Audio Commentary Datasetcited by 12
- 2023A Survey of Deep Learning in Sports Applications: Perception, Comprehension, and Decisioncited by 11
- 2024TeamTrack: A Dataset for Multi-Sport Multi-Object Tracking in Full-pitch Videoscited by 11
- 2025SoccerMaster: A Vision Foundation Model for Soccer Understandingcited by 10
- 2022Location analysis of players in UEFA EURO 2020 and 2022 using generalized valuation of defense by estimating probabilitiescited by 8
- 2023All Keypoints You Need: Detecting Arbitrary Keypoints on the Body of Triple, High, and Long Jump Athletescited by 8
- 2024Towards AI-Powered Video Assistant Referee System (VARS) for Association Footballcited by 8
- 2024SoccerNet 2024 Challenges Resultscited by 8
- 2025SoccerChat: Integrating Multimodal Data for Enhanced Soccer Game Understandingcited by 8
- 2023CLIP-ReIdent: Contrastive Training for Player Re-Identificationcited by 7
- 2023TAO-Amodal: A Benchmark for Tracking Any Object Amodallycited by 7
- 2024Foul prediction with estimated poses from soccer broadcast videocited by 7
- 2024Deep learning for action spotting in association football videoscited by 7
- 2025Domain Adaptation of VLM for Soccer Video Understandingcited by 7
- 2023Tracking Skiers from the Top to the Bottomcited by 6
- 2024Improving Object Detection Quality in Football Through Super-Resolution Techniquescited by 6
- 2025No Train Yet Gain: Towards Generic Multi-Object Tracking in Sports and Beyondcited by 6
- 2025FOOTPASS: A Multi-Modal Multi-Agent Tactical Context Dataset for Play-by-Play Action Spotting in Soccer Broadcast Videoscited by 5
- 2024Basketball-SORT: An Association Method for Complex Multi-object Occlusion Problems in Basketball Multi-object Trackingcited by 4
- 2024Engineering an Efficient Object Tracker for Non-Linear Motioncited by 4
- 2025TrackID3x3: A Dataset and Algorithm for Multi-Player Tracking with Identification and Pose Estimation in 3x3 Basketball Full-court Videoscited by 4
- 2025CAMELTrack: Context-Aware Multi-cue ExpLoitation for Online Multi-Object Trackingcited by 4
- 2025Deep Learning-Based Multi-Object Tracking: A Comprehensive Survey from Foundations to State-of-the-Artcited by 4
- 2023SoccerNet 2023 Tracking Challenge -- 3rd place MOT4MOT Team Technical Reportcited by 3
- 2024Enhanced Multi-Object Tracking Using Pose-based Virtual Markers in 3x3 Basketballcited by 3
- 2026GenTac: Generative Modeling and Forecasting of Soccer Tacticscited by 3
- 2024OSL-ActionSpotting: A Unified Library for Action Spotting in Sports Videoscited by 2
- 2025SoccerSynth-Detection: A Synthetic Dataset for Soccer Player Detectioncited by 2
- 2025Pixels or Positions? Benchmarking Modalities in Group Activity Recognitioncited by 2
- 2025RacketVision: A Multiple Racket Sports Benchmark for Unified Ball and Racket Analysiscited by 2
- 2023Runner re-identification from single-view running video in the open-world settingcited by 1
- 2023A Large-Scale Re-identification Analysis in Sporting Scenarios: the Betrayal of Reaching a Critical Pointcited by 1
- 2024Multi Player Tracking in Ice Hockey with Homographic Projectionscited by 1
50 most cited of 80 All 80 (JSON)
- 2026Training-Free Long-Term Multi-Object Tracking for Sports Video Analytics16 Aug 2026
- 2026BasketEvent: Understanding Who Did What and When in Basketball Videos23 Jul 2026
- 2026An LLM-Based Automatic Sportscast Solution for Robot Soccer Matches16 Jul 2026
- 2026TreeSoc: Tree-Structured Dynamic Reasoning and Tool Synergy for Soccer Video Understanding13 Jul 2026
- 2026WaspMOT: A Benchmark for Long-Term Multi-Object Tracking of Trichogramma Wasps9 Jul 2026
- 2026SoccerNet 2026 Challenges Results8 Jul 2026
- 2026AVTrack: Audio-Visual Tracking in Human-centric Complex Scenes1 Jun 2026
- 2026Policy-based Foveated Imaging and Perception1 Jun 2026
- 2026SVI-Bench: A Dynamic Microworld for Strategic Video Intelligence29 May 2026
- 2026GenTac: Generative Modeling and Forecasting of Soccer Tactics13 Apr 2026
- 2026Towards Athlete Fatigue Assessment from Association Football Videos7 Apr 2026
- 2026TacSIm: A Dataset and Benchmark for Football Tactical Style Imitation26 Mar 2026
38 more
- 2026GTATrack: Winner Solution to SoccerTrack 2025 with Deep-EIoU and Global Tracklet Association31 Jan 2026
- 2025Learning Association via Track-Detection Matching for Multi-Object Tracking26 Dec 2025
- 2025Multi-Modal Soccer Scene Analysis with Masked Pre-Training22 Dec 2025
- 2025Avatar4D: Synthesizing Domain-Specific 4D Humans for Real-World Pose Estimation18 Dec 2025
- 2025SoccerMaster: A Vision Foundation Model for Soccer Understanding11 Dec 2025
- 2025From Detection to Association: Learning Discriminative Object Embeddings for Multi-Object Tracking2 Dec 2025
- 2025RacketVision: A Multiple Racket Sports Benchmark for Unified Ball and Racket Analysis21 Nov 2025
- 2025Enhancing Multi-Camera Gymnast Tracking Through Domain Knowledge Integration20 Nov 2025
- 2025FOOTPASS: A Multi-Modal Multi-Agent Tactical Context Dataset for Play-by-Play Action Spotting in Soccer Broadcast Videos20 Nov 2025
- 2025Pixels or Positions? Benchmarking Modalities in Group Activity Recognition16 Nov 2025
- 2025Multi Camera Connected Vision System with Multi View Analytics: A Comprehensive Survey10 Oct 2025
- 2025Beyond the Individual: Introducing Group Intention Forecasting with SHOT Dataset25 Sep 2025
- 2025Motion Estimation for Multi-Object Tracking using KalmanNet with Semantic-Independent Encoding14 Sep 2025
- 2025TrajSV: A Trajectory-based Model for Sports Video Representations and Applications15 Aug 2025
- 2025MeMoSORT: Memory-Assisted Filtering and Motion-Adaptive Association Metric for Multi-Person Tracking13 Aug 2025
- 2025SoccerTrack v2: A Full-Pitch Panoramic Video Dataset for Game State Reconstruction and Ball Action Spotting3 Aug 2025
- 2025Deep Learning-Based Multi-Object Tracking: A Comprehensive Survey from Foundations to State-of-the-Art16 Jun 2025
- 2025No Train Yet Gain: Towards Generic Multi-Object Tracking in Sports and Beyond2 Jun 2025
- 2025SoccerChat: Integrating Multimodal Data for Enhanced Soccer Game Understanding22 May 2025
- 2025Domain Adaptation of VLM for Soccer Video Understanding20 May 2025
- 2025CAMELTrack: Context-Aware Multi-cue ExpLoitation for Online Multi-Object Tracking2 May 2025
- 2025TimeSoccer: An End-to-End Multimodal Large Language Model for Soccer Commentary Generation24 Apr 2025
- 2025TrackID3x3: A Dataset and Algorithm for Multi-Player Tracking with Identification and Pose Estimation in 3x3 Basketball Full-court Videos24 Mar 2025
- 2025Towards long-term player tracking with graph hierarchies and domain-specific features28 Feb 2025
- 2025SoccerSynth-Detection: A Synthetic Dataset for Soccer Player Detection16 Jan 2025
- 2024Enhanced Multi-Object Tracking Using Pose-based Virtual Markers in 3x3 Basketball9 Dec 2024
- 2024Towards Universal Soccer Video Understanding2 Dec 2024
- 2024Deep learning for action spotting in association football videos2 Oct 2024
- 2024Temporally Propagated Masks and Bounding Boxes: Combining the Best of Both Worlds for Multi-Object Tracking21 Sep 2024
- 2024SoccerNet 2024 Challenges Results16 Sep 2024
- 2024SynPlay: Large-Scale Synthetic Human Data with Real-World Diversity for Aerial-View Perception21 Aug 2024
- 2024Visible-Thermal Multiple Object Tracking: Large-scale Video Dataset and Progressive Fusion Approach2 Aug 2024
- 2024PLayerTV: Advanced Player Tracking and Identification for Automatic Soccer Highlight Clips22 Jul 2024
- 2024Towards AI-Powered Video Assistant Referee System (VARS) for Association Football17 Jul 2024
- 2024Investigating Event-Based Cameras for Video Frame Interpolation in Sports2 Jul 2024
- 2024OSL-ActionSpotting: A Unified Library for Action Spotting in Sports Videos1 Jul 2024
- 2024Engineering an Efficient Object Tracker for Non-Linear Motion30 Jun 2024
- 2024Basketball-SORT: An Association Method for Complex Multi-object Occlusion Problems in Basketball Multi-object Tracking28 Jun 2024
50 newest of 80 All 80 (JSON)
- 2023SportsMOT: A Large Multi-Object Tracking Dataset in Multiple Sports Scenescited by 131
- 2022Hard to Track Objects with Irregular Motions and Similar Appearances? Make It Easier by Buffering the Matching Spacecited by 66
- 2023Single-Model and Any-Modality for Video Object Trackingcited by 65
- 2023SoccerNet-Caption: Dense Video Captioning for Soccer Broadcasts Commentariescited by 57
- 2023VARS: Video Assistant Referee System for Automated Soccer Decision Making from Multiple Viewscited by 38
- 2022DeepSportradar-v1: Computer Vision Dataset for Sports Understanding with High Quality Annotationscited by 37
- 2024MatchTime: Towards Automatic Soccer Game Commentary Generationcited by 37
- 2023Iterative Scale-Up ExpansionIoU and Deep Features Association for Multi-Object Tracking in Sportscited by 29
- 2022SoccerNet 2022 Challenges Resultscited by 28
- 2024Towards Universal Soccer Video Understandingcited by 21
- 2023SoccerNet 2023 Challenges Resultscited by 20
- 2023Towards Active Learning for Action Spotting in Association Football Videoscited by 14
12 most cited of 75 All 75 (JSON)
- 2026Training-Free Long-Term Multi-Object Tracking for Sports Video Analytics16 Aug 2026
- 2026BasketEvent: Understanding Who Did What and When in Basketball Videos23 Jul 2026
- 2026An LLM-Based Automatic Sportscast Solution for Robot Soccer Matches16 Jul 2026
- 2026WaspMOT: A Benchmark for Long-Term Multi-Object Tracking of Trichogramma Wasps9 Jul 2026
- 2026SoccerNet 2026 Challenges Results8 Jul 2026
- 2026AVTrack: Audio-Visual Tracking in Human-centric Complex Scenes1 Jun 2026
- 2026Policy-based Foveated Imaging and Perception1 Jun 2026
- 2026GenTac: Generative Modeling and Forecasting of Soccer Tactics13 Apr 2026
- 2026Towards Athlete Fatigue Assessment from Association Football Videos7 Apr 2026
- 2026TacSIm: A Dataset and Benchmark for Football Tactical Style Imitation26 Mar 2026
- 2026GTATrack: Winner Solution to SoccerTrack 2025 with Deep-EIoU and Global Tracklet Association31 Jan 2026
- 2025Learning Association via Track-Detection Matching for Multi-Object Tracking26 Dec 2025
12 newest of 75 All 75 (JSON)