This is an old revision of the document!
Seminar: Foundational Models for 2D and 3D Computer Vision (5 ECTS)
Winter Semester 2024/25, TU München
Organisers: Dominik Schnaus, Tarun Yenamandra
Please direct questions to fmcv-ws23@cvpr.in.tum.de
Course Materials: Course Materials (password protected; please email us or look in the preliminary meeting slides for the password)
News
2024-06-24: [preliminary meeting] will take place from 14:00 - 15:00 on 27.06.2024 online via Zoom. The slides will be published on a protected page linked from here. Please attend the meeting to learn more about the course or ask questions. Attendance at the preliminary meeting is not mandatory. The zoom link can be found on TUMonline.
Course Description
Large-scale machine learning models, also called foundational models, have proven helpful in many applications. As they are learned from large amounts of data, they can be used as priors in many applications. Different machine learning paradigms, such as transformers and diffusion models, have been used to learn such foundational models. While the major focus has been on 2D foundational models, some methods have also been proposed to extend the 2D foundational models to 3D using differentiable rendering. This seminar will discuss some important foundational models in recent years and their applications.
Prerequisites
All participants should have a solid working knowledge of linear algebra and calculus. In addition, it is useful (but not required) for students to have a background in one of the following topics: computer vision, image processing, computer graphics, 3D geometry, and deep learning.