Contrastive Learning for Sports Video: Unsupervised Player Classification

Koshkina, Maria; Pidaparthy, Hemanth; Elder, James H.

doi:10.1109/cvprw53098.2021.00510

Cited by 24 publications

(12 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2011-2016 2017-present Football [37]- [42] [30], [43]- [52] Basketball [53]- [59] [60]- [72] Volleyball [73]- [77] [78]-[83] Hockey [84]- [89] [90]-[99] Diving [100] [101]-[107] Tennis [108]- [113] [114]- [123] Table tennis [124]- [129] [130]-[138] Gymnastics [139]- [144] [145]-[148] Badminton [149]- [154] [155]- [164] Figure Skating [165], [166] [2], [167]- [174] Recently, researchers in the communities of computer vision and sports pay much attention to sports video analysis, including building datasets and proposing novel methodologies [2], [17]- [30]. In most existing works on sports video analysis, recognizing the actions of players in videos is crucial.…”

Section: Sportmentioning

confidence: 99%

A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications

Wu¹,

Wang²,

Bian³

et al. 2022

Preprint

View full text Add to dashboard Cite

To understand human behaviors, action recognition based on videos is a common approach. Compared with imagebased action recognition, videos provide much more information. Reducing the ambiguity of actions and in the last decade, many works focused on datasets, novel models and learning approaches have improved video action recognition to a higher level. However, there are challenges and unsolved problems, in particular in sports analytics where data collection and labeling are more sophisticated, requiring sport professionals to annotate data. In addition, the actions could be extremely fast and it becomes difficult to recognize them. Moreover, in team sports like football and basketball, one action could involve multiple players, and to correctly recognize them, we need to analyse all players, which is relatively complicated. In this paper, we present a survey on video action recognition for sports analytics. We introduce more than ten types of sports, including team sports, such as football, basketball, volleyball, hockey and individual sports, such as figure skating, gymnastics, table tennis, tennis, diving and badminton. Then we compare numerous existing frameworks for sports analysis to present status quo of video action recognition in both team sports and individual sports. Finally, we discuss the challenges and unsolved problems in this area and to facilitate sports analytics, we develop a toolbox using PaddlePaddle 1 , which supports football, basketball, table tennis and figure skating action recognition.

show abstract

Section: Sportmentioning

confidence: 99%

A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications

Wu¹,

Wang²,

Bian³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In comparison with this method, our method does not rely on pixellevel annotations for training the network. Another recent approach [29] trains an embedding network for team player discrimination using a triplet loss for contrastive learning. Contrary to this approach, our method works on partial views of the field and not only clusters the two majority classes (players from team 1 and team 2) but also considers minority classes (referees and goalkeepers).…”

Section: Related Work Action Spotting and Video Summarization In Sportsmentioning

confidence: 99%

A Graph-Based Method for Soccer Action Spotting Using Unsupervised Player Classification

Cartas

Ballester

Haro

2022

Proceedings of the 5th International ACM Workshop on Multimedia Content Analysis in Sports

View full text Add to dashboard Cite

Action spotting in soccer videos is the task of identifying the specific time when a certain key action of the game occurs. Lately, it has received a large amount of attention and powerful methods have been introduced. Action spotting involves understanding the dynamics of the game, the complexity of events, and the variation of video sequences. Most approaches have focused on the latter, given that their models exploit the global visual features of the sequences. In this work, we focus on the former by (a) identifying and representing the players, referees, and goalkeepers as nodes in a graph, and by (b) modeling their temporal interactions as sequences of graphs. For the player identification, or player classification task, we obtain an accuracy of 97.72% in our annotated benchmark. For the action spotting task, our method obtains an overall performance of 57.83% average-mAP by combining it with other audiovisual modalities. This performance surpasses similar graph-based methods and has competitive results with heavy computing methods. Code and data are available at https://github. com/IPCV/soccer_action_spotting.

show abstract

“…Early work used colour histograms or colour features with a clustering approach to differentiate between teams [1,3,7,13,16,23,30,32,34,44]. This approach, while being lightweight, does not handle occlusions, changes in illumination, and teams wearing similar jersey colours well [3,25]. Deep learning approaches have increased performance and generalizablitity of player classification models [22].…”

Section: Team Identificationmentioning

confidence: 99%

“…Finally, the player's team identification is determined by the channel that contains the maximum proportions of pixels. Koshkina et al [25] use contrastive learning to classify player bounding boxes in hockey games. This self-supervised learning approach uses a CNN trained with triplet loss to learn a feature space that best separates players into two teams.…”

Section: Team Identificationmentioning

confidence: 99%

Player Tracking and Identification in Ice Hockey

Vats¹,

Walters²,

Fani³

et al. 2021

Preprint

View full text Add to dashboard Cite

Tracking and identifying players is a fundamental step in computer vision-based ice hockey analytics. The data generated by tracking is used in many other downstream tasks, such as game event detection and game strategy analysis. Player tracking and identification is a challenging problem since the motion of players in hockey is fast-paced and non-linear when compared to pedestrians. There is also significant camera panning and zooming in hockey broadcast video. Identifying players in ice hockey is challenging since the players of the same team look almost identical, with the jersey number the only discriminating factor between players. In this paper, an automated system to track and identify players in broadcast NHL hockey videos is introduced. The system is composed of three components (1) Player tracking, (2) Team identification and (3) Player identification. Due to the absence of publicly available datasets, the datasets used to train the three components are annotated manually. Player tracking is performed with the help of a state of the art tracking algorithm obtaining a Multi-Object Tracking Accuracy (MOTA) score of 94.5%. For team identification, the away-team jerseys are grouped into a single class and hometeam jerseys are grouped in classes according to their jersey color. A convolutional neural network is then trained on the team identification dataset. The team identification network gets an accuracy of 97% on the test set. A novel player identification model is introduced that utilizes a temporal one-dimensional convolutional network to identify players from player bounding box sequences. The player identification model further takes advantage of the available NHL game roster data to obtain a player identification accuracy of 83%.

show abstract

Contrastive Learning for Sports Video: Unsupervised Player Classification

Cited by 24 publications

References 44 publications

A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications

A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications

A Graph-Based Method for Soccer Action Spotting Using Unsupervised Player Classification

Player Tracking and Identification in Ice Hockey

Contact Info

Product

Resources

About