
STEPHEN P. KAPUTSOS
XR & AI/CV ENGINEER

Photorealistic and Digital Twin Sports Simulator for NBA & NFL
Leveraging the benefits of virtual environments to simulate venues for sports analytics
ROLE: Senior Computer Vision & Unreal Engine Engineer / Researcher
SKILLS UTILIZED: Procedural Houdini generation, Environment Design & Development, 3D Modeling & Animation (Maya), Data Collection, Data Processing, Data Analysis, Coding (C++, Python)
SOFTWARE: Unreal Engine, Houdini, Substance Painter & Designer, and Maya
ABOUT: Sports such as basketball and football can greatly benefit from AI and computer vision systems. These technologies have the potential to revolutionize sports analytics and foster a new degree of athlete performance. However, in order to apply AI/CV to sports such as these, a photorealistic simulator is required. Without such a medium for testing algorithms, performance discrepancies between the virtual and physical worlds can not be discerned.
​
Photorealistic and digital twin venues were developed using Unreal Engine 4/5 for the NBA and NFL to serve as an AI/CV test-bed and real-time game mirror for fans. All environments are coupled with an XR exploration mode that is compatible with popular AR/VR headsets. The sports simulator allows for full customization of athlete avatars, team setup, and analytic report export. This is an invaluable tool on the road to a new generation of sports analytics.


