Motion Capture at Gamenic: Bringing Digital Characters to Life

Character animation is one of the most challenging disciplines in digital production. Even subtle imperfections in movement can trigger the uncanny valley effect, breaking the illusion of a living character. At Gamenic, we use motion capture technology to bridge the gap between digital puppetry and believable human performance.

Our motion capture setup has evolved significantly over the past year. We started with a traditional marker-based system using OptiTrack cameras, which remains our gold standard for precision work. More recently, we’ve added markerless AI-based tracking using computer vision, which offers faster setup times and more natural performer movement at the cost of slightly reduced accuracy.

The raw capture data is just the beginning. Our animation team spends considerable time cleaning, retargeting, and enhancing mocap data. Cleanup involves removing noise and correcting tracking errors. Retargeting maps the human performer’s proportions to the digital character’s often non-human anatomy. Enhancement adds secondary animation cloth simulation, hair dynamics, facial muscle detail that mocap alone can’t provide.

Facial capture deserves special attention. We use a combination of Apple’s ARKit-based face tracking for rapid iteration and a dedicated Faceware system for hero shots requiring maximum fidelity. Our facial animation pipeline includes a custom blend shape library of over 200 expressions, allowing nuanced performances that convey subtle emotions.

Virtual production is increasingly becoming part of our mocap workflow. Using Unreal Engine’s Live Link system, directors can see their actors’ performances applied to digital characters in real-time, composited into virtual environments on large LED volumes. This instant visual feedback transforms the creative process, allowing performance direction that was previously impossible in animated productions.