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.

No abstract

PurposeTo accurately provide evaluations on how match performance for elite skaters in short track speed skating developed, and whether geographical factors of ice rink locations should be considered apart from technical abilities. We created a dataset containing competition records from the 2013–14 to 2020–21 seasons (500 m event) on the official website.MethodsOne-way ANOVA was applied to statistically analyze whether the best performance times exhibited significant differences in varied hosting cities. Performance–time matrix and multivariate regression model were further established to quantitatively explain how geographical factors (longitude, latitude, altitude, and barometric pressure) affected performance.ResultsOur findings firstly confirmed that the fastest 500 m finishing times varied due to the hosting cities (P = 0.008) and showed that venue locations could boost or impair performance time with the maximum range of 3.6 s. Meanwhile, latitude (slightly over 46° when performance is maximized) was the most influential factor to account for the performance–time difference in different ice rink locations according to the multivariate regression model, though altitude (1,225 meters when performance is maximized) was also important.ConclusionsIn this perspective, elite skaters should check the geographical factors of the venues before they participated in the upcoming competitions, assess the real strength of their rivals, and adopt flexible tactics during training sessions.

No abstract

No abstract