Login
Block Seminar: Deep-learning-based 3D scene representation and understanding for mobile robots
Seminar Description
The representation and understanding of the surroundings are the prerequisites for autonomous intelligent robots. The focus of this seminar is investigating the recent advances in enhancing the scene representation and understanding of mobile robots, especially in challenging environments with the interference of dynamic objects, time-variant appearance, poor illuminations, and bad weather conditions, etc. Students are required to investigate the specific topics by referring to the relevant papers and to present their findings at a colloquium in an understandable way. The related topics include but are not limited to:
- Scene and object representation
- Neural implicit representation
- Scene graph
- Pose estimation of mobile robots
- 3D reconstruction
- Semantic segmentation
- Pose and shape estimation of dynamic objects
- Long-term autonomy
- 3D Scene understanding
- Semantic exploration
Organization
General
- Organizers: Dr. Xingxing Zuo
- E-Mail: xingxing.zuo@tum.de
- Due to the current situation with the Coronavirus, the preliminary meeting will be online. The seminar can be in person but stay tuned.
Preliminary Meeting
- Date: Jan. 31, 2022
- Time: 14:00 - 15:00
- Location: Zoom
Formal Seminar Days
- Time:
- 14:00 - 17:00, Monday, May. 30, 2022
- 09:00 - 12:00, 14:00 - 17:00, Tuesday, May. 31, 2022
- Location: MI 01.07.023 (Monday); MI 01.13.007 (Tuesday)
- Talks will be held in English
Material
All course-related material are uploaded on this page (password required).
