Presentation given by Daniel Cremers on 22nd February 2023 in the one world seminar on the mathematics of machine learning on the topic "Self-Supervised Learning for 3D Shape Analysis".
Abstract: While neural networks have swept the field of computer vision and are replacing classical methods in many areas of image analysis and beyond, extending their power to the domain of 3D shape analysis remains an important open challenge. In my presentation, I will focus on the problems of shape matching, correspondence estimation and shape interpolation and develop suitable deep learning approaches to tackle these challenges. In particular, I will focus on the difficult problem of computing correspondence and interpolation for pairs of shapes from different classes -- say a human and a horse -- where traditional isometry assumptions no longer hold.
4 июл 2024