Detection of Ice Hockey Players and Teams via a Two-Phase Cascaded CNN Model

Abstract

The accurate detection of ice hockey players and teams during a game is crucial to the tracking of individual players on the rink and team tactical decision making and is therefore becoming an important task for coaches and other analysts. However, hockey is a fluid sport due to its complex situation and the frequent substitutions by both teams, resulting in the players taking various postures during a game. Few player detection models from basketball and soccer take these characteristics into account, especially for team detection without prior annotations. Here, a two-phase cascaded convolutional neural network (CNN) model is designed for the detection of individual ice hockey players, and the jersey color of the detected players is extracted to further identify team affiliations. Our model filters most of the disturbing information, such as the audience and sideline advertising bars, in Phase I and refines the detection of the targeted players in Phase II, resulting in an accurate detection with a precision of 98.75% and a recall of 94.11% for individual players and an average accuracy of 93.05% for team classification with a self-built dataset of collected images from the 2018 Winter Olympics. The results for the regular season games of the 2019-2020 National Hockey League (NHL) covering all 31 teams are also presented to show the robustness of our model. Compared to state-of-the-art approaches, our player detection model achieves the highest accuracy with the self-built dataset.

Estimation of Heart Rate Using Regression Models and Artificial Neural Network in Middle-Aged Adults

“…The model parameters were updated during the minimization of the loss function through the forward propagation and back propagation phases. ANNs and other computational models are popular for predicting values in other sports science fields, such as player detection ( Guo et al, 2020 ) and the investigation of exercise-mediated diseases ( Tao et al, 2021 ), owing to their outstanding abilities for generalization and efficiency in investigating nonlinear latent relations between variables. During model establishment, four sets of variable combinations were verified based on the stepwise variable selection criteria ( Figure 1 ).…”

Champion Position Analysis in Short Track Speed Skating Competitions From 2007 to 2019

“…The use of artificial intelligence technologies to visualize and analyze data gains favor in modern sports science ( Guo et al, 2020 ), due to its outstanding abilities for generalization and efficiency in investigation of the non-linear latent interactions between variables. We accept that some limitations exist in the present study, such as the involvement of just starting and finishing positions instead of incorporating more detailed information gathered from race profiles (i.e., split time).…”

A Feature-Weighting Approach Using Metaheuristic Algorithms to Evaluate the Performance of Handball Goalkeepers

“…To address problem (5), the major concern or first step is to determine the codification of the solutions for the optimization problem. There must be a simple codification that does not increase the computational cost in solving the problem, and is versatile in such a way that the coding does not change regardless of the algorithm or the resolution method employed.…”

“…The exploration and exploitation trade-off is controlled by both the G t and Kbest t functions. An overview of the pseudocode of GSA is shown in Algorithm (5). The justification for choosing GSA as a reference optimization algorithm for the experiments made in this paper is based on its good performance in a multitude of domains and on the recent application in the selection and estimation of parameters for the selection of players in the call-up and in the line-up in a match [53], [54].…”

“…In handball, although the scientific and professional attention given to quantitative data analysis has grown very rapidly over the past few years, little work has been done on evaluating a player in a game-by-game scenario [3]. Thus, handball is a long way behind the analysis existing in other sports, such as soccer [4] or ice hockey [5]. The existing approaches [6] used to represent the effectiveness of a handball team or a single handball player is now insufficient for evaluating the performance of players in a match.…”