Breaking News

These are the 20 best travel destinations for summer 2024, according to Google Flight Searches 3 Google Maps updates to make summer travel easier SPACECENT is up the new war zone > United States Space Force > Article Display Tuberculosis — United States, 2023 | MMWR Thousands of US bridges are vulnerable to collapse from a single hit: NTSB Why don’t the Blazers or ROOT Sports offer standalone streaming? Up to 200,000 people estimated to travel to Vermont for total solar eclipse How fast will April’s total solar eclipse travel? The UN Security Council demands a ceasefire in Gaza during Ramadan Mexico in the emerging world order

Algorithms developed in Cornell’s Laboratory of Intelligent Systems and Controls can predict the in-game actions of volleyball players with more than 80% accuracy, and now the lab is working with the Big Red hockey team to expand the use of the research project.

The algorithms are unique in that they take a holistic approach to action management, combining visual data – for example, when a player is on the court – with specific information, such as the player’s exact role on the team.

“Computer vision can interpret visual information such as the color of the jersey and the player’s position or body shape,” said Silvia Ferrari, John Brancaccio Professor of Mechanical and Aerospace Engineering, who led the research. “We still use real-time information, but we include hidden things like team strategy and player roles, things that we as humans can talk about because we are experts at this point.”

Doctoral student Junyi Dong works with his colleagues and other doctoral students in their laboratory in Upson Hall.

Ferrari and doctoral students Junyi Dong and Qingze Huo trained algorithms to predict hidden variables in the same way that people get sports information – by watching sports. The algorithms used machine learning to extract data from videos of volleyball games, and then used that data to help predict when shown a new set of games.

The results were released Sept. 22 in the journal ACM Transactions on Intelligent Systems and Technology, and shows algorithms that can affect the work of players – for example, to distinguish a passer from a blocker – with an average accuracy of about 85%, and can predict many actions over a sequence of up to 44 frames with an average accuracy of more than 80 %. The activities included scratching, setting, blocking, digging, running, scratching, falling, standing and jumping.

Ferrari sees teams using algorithms to better prepare for competition by training them with the opponent’s current game and using their predictive abilities to learn specific games and game situations.

Ferrari has filed for a patent and is now working with the Big Red men’s hockey team to further develop the software. Using a game plan provided by the team, Ferrari and his graduate students, led by Frank Kim, are developing algorithms that identify players, actions and game situations. Another goal of the project is to help define game film, which is a tedious task when done manually by team members.

“Our program puts a lot of emphasis on video analysis and data technology,” said Ben Russell, director of hockey operations for the Cornell men’s team. “We are always looking for ways to improve as a coaching staff so that we can serve our players better. I have been very impressed with the research done by Professor Ferrari and his students so far. I believe this program has the potential to dramatically influence the way teams learn and prepare for competition. “

Outside of sports, the ability to anticipate human actions has great potential for the future of human-machine interaction, according to Ferrari, who said that developed software can help autonomous vehicles make better decisions, bring robots and humans closer together in warehouses, and they can even design. video games are fun for increasing the computer’s intelligence.

“People are not as imaginative as machine learning algorithms are making them now,” said Ferrari, associate professor of online-campus engineering research, “because if you really think about all of the content, all of the content, and you see a group of people, you can get better at predicting what they’re going to do.” “

The research was supported by the Office of Naval Research Code 311 and Code 351, and the commercial effort is being supported by the Cornell Office of Technology Licensing.

Syl Kacapyr is associate director of marketing and communications for the College of Engineering.

Leave a Reply

Your email address will not be published. Required fields are marked *