Technical Name KarmaZone Strike Zone3D Markerless Motion Analysis System
Project Operator National Taiwan University of Sport
Project Host 吳育任
Six synchronized high-speed cameras were installed in the baseball stadium in NTUS. We established the calibration equation of the multi-camera systemestablished an electronic strike zone system, KarmaZone, that can automatically coordinates when the pitch crossed the home plate. From the slow motion videos recorded, the markerless 3D human posture estimation has been developed with the deep 
Scientific Breakthrough
A baseball player's motion is captured by 6 high speed camerasanalyzed using deep learning models. We used YOLOv3, AlphaPose, k-means clustering, triangulation,OpenCV to achieve the goal. Our goal is to run it real time during the game so both computation timeaccuracy is very important. We use CNN to trainoptimize multiple cameras in the baseball stadium. Using OpenCV to increa
Industrial Applicability
We use high speed camera to construct electric strike zone to accurate strike/ball calls. The footage captured is also used for markerless motion analysis to improve the moving efficiencykinetic chain of the player. Along with the pitch-by-pitch scoring system, the KarmaZone system provides all the tools needed for Baseball Analyticstraining.
Keyword baseball strike zone baseball K-zone markerless motion analysis motion analysis sabermetrics sport science precision sports motor learning motion analysis baseball trajectory reconstruction
  • Contact