392153 Neural and Geometric Modeling (S) (WiSe 2024/2025)

Three-dimensional geometric models are the base for applications in computer graphics, computer vision, computer-aided design, and many other related fields. This course will address computerized modeling of 3D geometry using

• point clouds, triangles, and polygonal meshes, the classical 3D shape representation
• implicit representations, which require a solve step to recover explicit geometry
• neural representations, which use neural networks to encode geometry

Central will be data structures and algorithms for creating, manipulating, editing, and analyzing 3D models

The course is largely based on courses given at UBC and SFU:
https://www.cs.ubc.ca/~sheffa/dgp/
https://www.sfu.ca/outlines.html?2023/spring/cmpt/764/g100

The following slides provide additional details and context to the existing visual computing modules at Bielefeld University by the Visual AI group:
https://docs.google.com/presentation/d/e/2PACX-1vRGyE0j_cU324qp-vOEE4hEBPn3HdwliZQqeg_vT0w7Yt9EUXmgFB6NamL_ynqeBMYRgD-cdvfvXobA/pub?start=false&loop=false&delayms=3000

Teilnahmevoraussetzungen, notwendige Vorkenntnisse

Required skills: Excellent Python programming skills (>1 year), fundamental mathematics (linear algebra), and basics of machine learning or neural networks.
Optional skills: PyTorch or C++ experience and knowledge in one of geometry, computer graphics, or computer vision (lectures or projects) helps but is not required.

Students need not be experts in all aspects of visual computing but have a strong background in one of graphics, vision, and machine learning. These skills are typically acquired through lectures or by private projects, hackathons, and coding competitions.

Literaturangaben

The papers that will be discussed are from the CVPR, SIGGRAPH, and NeurIPS conferences, or other respected computer vision and graphics venues.

The basis for the lectures is the following book:
"Polygon Mesh Processing" by Mario Botsch, Leif Kobbelt, Mark Pauly, Pierre Alliez, Bruno Levy. October 7, 2010 by A.K. Peters/CRC Press.
ISBN: 9781568814261

An excellent practical resource is the polyscope documentation that provides mathematical reasoning and interactive visualization (and is recently available as a python package):
https://polyscope.run/py/

To download models to test your code on, check aim@shape or Thingi10k. You always want to do initial testing on simple surfaces such as a sphere, a cube, or a plane which you can create/export and view using common tools, such as MeshLab, or polyscope.

Externe Kommentarseite

https://docs.google.com/presentation/d/e/2PACX-1vRGyE0j_cU324qp-vOEE4hEBPn3HdwliZQqeg_vT0w7Yt9EUXmgFB6NamL_ynqeBMYRgD-cdvfvXobA/pub?start=false&loop=false&delayms=3000