Manifold correspondence: a signal processing perspective

Опубликовано 22 июня 2016, 1:54

In recent years, geometric data is gaining increasing interest both in the academia and industry. In computer graphics and vision, this interest is owed to the rapid development of 3D acquisition and printing technologies, as well as the explosive growth of publicly-available 3D shape repositories. In machine learning, there is a gradual understanding that geometric structure plays an important role in high-dimensional complicated datasets. In this talk, I will use the problem of manifold correspondence (a fundamental and notoriously hard problem with a wide range of applications in geometric processing, graphics, vision, and learning) as a showcase for classical methods from the domain of signal processing (such as sparse coding, joint diagonalization, and matrix completion) applied to geometric problems. I will show applications to 3D shape correspondence, multi-view clustering, and image labelling.