Giulia Martinelli

I am a Post-Doctoral Researcher at the University of Trento, working at the intersection of Computer Vision and Computer Graphics. I collaborate with researchers across academia and industry on advancing the full pipeline of digital human creation—from high-fidelity 3D appearance reconstruction to motion capture, retargeting, and real-time animation. My work focuses on developing computational models that capture, understand, and synthesize realistic human shape, motion, and behavior, enabling their integration into interactive virtual environments.

More specifically, I’m interested in:

• Digital human modeling and parametric representations, including SMPL/SMPL-X–based methods for capturing shape, expression, and articulation.

• 3D human motion capture and reconstruction, spanning markerless/* marker-based pipelines, volumetric capture, and robust pose estimation.

• Motion retargeting and animation, with a focus on automatic rigging, skinning, and real-time control for XR and virtual characters.

• Applications in sports analytics, AR/VR, and embodied AI, enabling accurate analysis, simulation, and interaction of human motion in diverse environments.