With the advance of modern technology and computer power, processing and analysis of 3D shapes becomes a ubiquitous task due to fast acquisition and frequent use of 3D data. Recent developments of machine learning bring many state-of-the-art results in image processing. However, extension of deep learning methods in 3D shape analysis is still in an early state due to the challenge of non-Euclidean structures of 3D shapes. The mini symposium aims at enhancing interaction of scholars working on learning methods for 3D shape analysis.

Learning Geometry

Ron Kimmel (Technion - Israel Institute of Technology)

Geometric Interpretation to GAN model

Xianfeng Gu (State University of New York at Stony Brook )

Tradeoffs between speed and accuracy in inverse problems

Alexander Bronstein (Technion - Israel Institute of Technology)

The Geometry of Synchronization Problems and Learning Group Actions

Tingran Gao (The University of Chicago)

Organizers:

Rongjie Lai (Rensselaer Polytechnic Institute)

Ronald Lui (Chinese University of Hong Kong)

Keywords:

computer graphics, computer vision, deep learning, machine learning, nonlinear optimization