Video-Based Performance Evaluation for ECR Drills

This paper addresses the challenge of automatically assessing performance in military training exercises (ECR drills) within synthetic environments. It proposes a video-based system that uses computer vision to extract data (skeletons, gaze, trajectories) and derive metrics for psychomotor skills, situational awareness, and teamwork. This approach offers a less intrusive and potentially more scalable alternative to traditional methods, providing actionable insights for after-action reviews and feedback.