Nicola Garau

I am an Assistant Professor at the University of Trento, working on computer graphics and inverse graphics. I am the current holder of the Computer Graphics masters' course.

I earned my PhD in Computer Vision from the University of Trento, under the supervision of Nicola Conci, where I worked on viewpoint-equivariant neural networks for human pose estimation and brain-inspired models for interpretable representation learning. Earlier, I was an intern at the Perceptual Computing Lab at Carnegie Mellon University, working on OpenPose, and later at Epic Games, under the supervision of Carsten Stoll, Denis Tomè and Iain Matthews, working on monocular unsupervised 3D human pose estimation in the wild.

My current research connects real-time graphics, synthetic data, human motion, and learning inside 3D environments. I was involved in a national project at the ICSC National Centre for HPC, Big Data and Quantum Computing, where I worked on large-scale urban digital twins, and I developed Unreal Engine-based pipelines such as To Infinity and Beyond for GPU-driven synthetic data generation and processing.

My work has appeared at venues including CVPR, ICCV, ECCV, SIGGRAPH, and Eurographics. I am interested in building graphics systems that do not only render the world, but also simulate, measure, and understand it for machine learning and inverse graphics.