Project lead says, the technology could lead to major sport changes, as the clubs will identify talented players and quickly recruit them. The current player performance analysis involves someone who watches videotapes of matches and manually logs individual player actions, recording how numerous shots and passes a player took, where the action was taken and whether it had a successful outcome.
Some automated technologies are already on the market but can only track players on the pitch - to determine the covered distance and speed - but not provide detailed data on players' action.
Dr. Li and her team were working on a hybrid systems to accelerate human data entry to complement the high demand for cost-effective and timely performance data from large amount of football videos. To address these issues, Dr. Li and her team have developed
Dr. Li and his team have been using an AI model to detect player body limbs and positions, in accordance with current developments in AI and deep learning, in order to identify and analyse their movements.
The technology processes video images, detects players and identifies whether they run, walk, or jump and with which foot they pass the ball. In order to do that, researchers have used in-depth learning (a new state-of-the-art machine learning technology) and computer vision.
In addition to examining actions in a game, Dr. Shreedhar Rangappa, a Research Associate who works on a project, trained the deep neural network to track players and gather data on the performance of individual players during the match video.
"The technology developed will make it possible to interpret the game far more objectively, as it underlines players' ability and team cooperation. "There will be a positive effect on football and on advancing sporting technology while giving value to the players , coaches and recruiters using the information."
Statmetrix has supported a prestigious MSDUK Innovation Challenge Award 2019 through its collaboration with Loughborough University and technology.
we hope you're doing well if you see any inappropriate phrases please let us know on our contact page at the bottom. thank you! .