Abstract
With the explosive growth of sports video data on the internet platform, how to scientifically manage this information has become a major challenge in the current big data era. In this context, a new lightweight player segmentation algorithm is proposed to realize the automatic analysis of basketball game video. Firstly, semantic events are expressed by extracting group and global motion features. A complete basketball game video is divided into three stages, and a basketball event classification method integrating global group motion patterns and domain knowledge is proposed. Secondly, a player segmentation algorithm based on lightweight deep learning is proposed to detect basketball players, segment the players, and finally extract players' spatial features based on deep learning to realize players' pose estimation. As the experimental results indicate, when a proposed 2-stage classification algorithm is used to classify the videos, the accuracy of identifying layup, the shooting, and other 2-pointers are improved by 21.26% and 6.41%, respectively. And the accuracy of average events sees an improvement of 2.74%. The results imply that the 2-stage classification based on event-occ is effective. After comparing the four methods of classifying players, it is found that there is no significant difference among these four methods about the accuracy of segmenting. Nevertheless, when judged with the time that these methods take separately, FCN-CNN (Fully Convolutional Network-Convolutional Neural Network) based on superpixels has overwhelming advantages. The event analysis method of basketball game video proposed here can realize the automatic analysis of basketball video, which is beneficial to promoting the rapid development of basketball and even sports.
Highlights
E spatial features of global motion patterns are extracted based on Convolutional Neural Network (CNN), and the player area is segmented based on the Fully Convolutional Network- Convolutional Neural Network (FCNCNN) method of superpixel clustering, to identify the player’s posture. e proposed FCN-CNN player segmentation algorithm based on superpixel filters out the complex background around the player, which is more conducive to subsequent attitude estimation
In the test stage of Superpixel-Convolutional Neural Network (SP-CNN), if five key points are selected at equal intervals for each superpixel, it takes 0.2 seconds to process a superpixel
In the test stage, it takes 20 seconds to set the number of superpixels of players to 100; that is, it takes 20 seconds to segment through SP-CNN
Summary
The technical revolution is booming. As the vital driving force of this industrial revolution, computer network technology is constantly changing people’s way of life and work, pushing human society into an intelligent new one with human-machine fusion and mutual sharing the creating. Ravi et al analyzed a single shooting video based on a deep learning algorithm. At first, they used VGG (Visual Geometry Group)-16 to extract the spatial features of movements information and sampling frames. E proposed FCN-CNN player segmentation algorithm based on superpixel filters out the complex background around the player, which is more conducive to subsequent attitude estimation. It realizes the automatic analysis of basketball video, assisting the coach to formulate tactics, the players to analyze actions, and the video viewers to quickly search interested video segments, and promoting the rapid development of basketball and even sports
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have