Oral or Written (depending on the number of students)

A virtual human is a digital representation of a real human. Virtual humans (VH) should look, move and eventually think like real humans. Building such VH is one of the long standing goals of Artificial Intelligence.
Learning them requires techniques and algorithms at the intersection of Machine Learning, Computer Vision and Computer Graphics. In this course, we will cover the key mathematical foundations and computational tools to learn VH from 3D scans, images and video of real humans. The course will cover classical representations of humans based on 3D meshes and textures, as well as modern ones where the appearance and behavior of virtual humans are encoded in neural networks.

Understand the mathematical tools and algorithms to build VH from data. At the end of the course, students will be familiar with the state of the art in human motion and shape modeling, estimation of pose, shape and humans in clothing from images and video, as well as learning generative models of human motion conditioned on 3D scene geometry. Students should be able to apply the concepts in practice, develop and train models, reproduce research and conduct original research in this area.

Related literature will be posted on the course website (https://virtualhumans.mpi-inf.mpg.de/DH22/). Knowledge of linear algebra, optimization and probability (e.g, mathematics for machine learning) and coding skills (Python).
Experience with deep learning (e.g, Deep Learning course).