Software Brings Sport Analytics Up to Speed
Jeffrey Heimgartner posted on May 10, 2018 | | 708 views

The world is increasingly becoming “smart”—smart appliances, smart cars, smart cities. Smart sports? Well, maybe not quite, but Rice University students have developed software that can provide automated sports analytics to potentially enable volleyball coaches and players to make smarter plays during their next games.

The student-designed software, Cherrypick, automatically analyzes volleyball matches and provides analytics of a game within one hour. The team chose volleyball due to the nature of the sport, which has specific stop and start points. Additionally, it’s a sport that often has teams playing games within a day or two of each other, making the need for quick data crucial for coaches.

“Cherrypick will automatically go through the game and extract the important statistical information and tie it directly to the video,” said James Grinage, electrical engineering major at Rice University. “Coaches can then easily go through the video and know when different plays took place and who was responsible for specific plays.”

Rice University’s students developed Cherrypick, a new sports analytics software. (Image by Jeff Fitlow, courtesy of Rice University.)
Rice University’s students developed Cherrypick, a new sports analytics software. (Image by Jeff Fitlow, courtesy of Rice University.)

Cherrypick’s ability to provide quick analytics sets it apart from other programs, which often require a wait time of 12 to 24 hours for the date  and often need to be completed manually. Coaches using Cherrypick can record a game, upload a video and receive statistics about the game within an hour thanks to the machine learning and computer vision algorithms developed by the students. Coaches can then make data-driven decisions based on plays, drills to practice, players to use and so on.

The team tested the accuracy of its software by tagging the ball position in dozens of hours of Rice and club volleyball matches. The students estimate they tagged 20 matches, which could last around two hours each, filmed at 30 frames per second.

“To the best of our knowledge, it’s the largest data set of volleyball tracking play segmentation in existence,” said Connor Heggie, electrical engineering major at Rice. “We’ve spent a good deal of time over the past year hand-tagging every single frame of video to show where the ball is at all times.”

To see the software in action, check out the video below:

The students plan to turn their idea into a startup company and expand its capabilities to potentially include tennis and baseball.

Interested in more student innovations? Check out Remote-Controlled Amphibious Vehicle Wades into the Surf and Robotic Spine Promises New Treatments for Spinal Deformities.

Recommended For You