Mentor/s
Jonathan Hudak
Participation Type
Poster
Abstract
Markerless motion capture systems offer promising data for analyzing athletic performance with high precision; however, the camera systems utilized as part of the Pioneer Performance Center have not been tested with higher speed motions. Theia 3D software, mainly used for gait analyses previously, was utilized to examine the effectiveness of the SONY RX 0II camera system and consistency of identifying key event markers during fast moving motions, such as pitching.
A specialized pipeline was developed to auto detect the frames corresponding to 4 different important event markers: maximum hip flexion, foot touchdown, late cocking, and ball release. The automated data generated by this pipeline was compared with manually identified event frames to assess both accuracy and precision.
Videos were captured at a rate of 120 frames per second, which is slower than other facilities collecting baseball kinematics. Notably, only fast balls were thrown during data collection sessions, with an average velocity of 75.9 miles/hour. No pitches were thrown with maximum effort due to the time of year the data was recorded.
The groundwork laid thus far has provided a solid foundation, regarding the pipeline. As more pitchers work with the Pioneer Performance Center, the pipeline will undergo further refinement and optimization to accommodate a diverse range of pitching styles and techniques, ensuring its effectiveness for all participants. Ultimately, the purpose of this study is to internally validate the data and pipeline to be used with more baseball players in the future.
College and Major available
Exercise Science BS
Location
Digital Commons & West Campus West Building University Commons
Start Day/Time
4-26-2024 12:00 PM
End Day/Time
4-26-2024 2:00 PM
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License
Prize Categories
Best Multidisciplinary Research or Collaboration, Most Scholarly Impact or Potential, Best Visuals
Can markerless motion capture software consistently and reliably auto detect important pitch event markers for baseball players?
Digital Commons & West Campus West Building University Commons
Markerless motion capture systems offer promising data for analyzing athletic performance with high precision; however, the camera systems utilized as part of the Pioneer Performance Center have not been tested with higher speed motions. Theia 3D software, mainly used for gait analyses previously, was utilized to examine the effectiveness of the SONY RX 0II camera system and consistency of identifying key event markers during fast moving motions, such as pitching.
A specialized pipeline was developed to auto detect the frames corresponding to 4 different important event markers: maximum hip flexion, foot touchdown, late cocking, and ball release. The automated data generated by this pipeline was compared with manually identified event frames to assess both accuracy and precision.
Videos were captured at a rate of 120 frames per second, which is slower than other facilities collecting baseball kinematics. Notably, only fast balls were thrown during data collection sessions, with an average velocity of 75.9 miles/hour. No pitches were thrown with maximum effort due to the time of year the data was recorded.
The groundwork laid thus far has provided a solid foundation, regarding the pipeline. As more pitchers work with the Pioneer Performance Center, the pipeline will undergo further refinement and optimization to accommodate a diverse range of pitching styles and techniques, ensuring its effectiveness for all participants. Ultimately, the purpose of this study is to internally validate the data and pipeline to be used with more baseball players in the future.
Students' Information
Jenna Engels: (honors) exercise science, graduating May 2024
Sarah Mandelkow: (honors) exercise science, graduating May 2024
Jacqueline Lage: (not honors) exercise science, graduating May 2024