3D Shape Analysis

Seminar – Summer Semester 2018

Organisers: Florian Bernard, Christian Theobalt

Organisation  |  Topics  |  Format  |  Format (Detail)  |  Resources

Functional Maps: A Flexible Representation of Maps Between Shapes (Ovsjanikov et al., SIGGRAPH 2012)

Overview

The objective of shape analysis is to recognise or quantify morphological patterns or regularities of some object(s) of interest. These insights may then help in developing theories and in better understanding the object class at hand. From a computational perspective, shape analysis is based on the fields computer science and (applied) mathematics, with diverse application areas such as computer vision, image analysis, computer graphics, computational geometry, or pattern recognition.

In this course we will cover various topics that are related to shape analysis. While there are several shape representations, e.g. point-clouds, surface meshes, or volumetric representations, a common aspect is the correspondence problem. Given two (or more) shapes, the correspondence problem refers to the task of bringing parts of the shapes into some meaningful correspondence, where the particular interpretation of ''meaningful'' depends on the application.

In this seminar we will cover recent research papers as well as well-established research papers from the following problem fields:

• shape deformation models,
• shape registration and matching,
• spectral shape analysis,
• geometric deep learning, and
• functional maps.

Topics

The following table lists more topics than available weeks – this is to allow some flexibility for the interests of the group to change the course structure. Once every participant has submitted their choice of topics, this list will be updated to show the presenter of each topic.