Segmentation, Reconstruction, Computer Vision
In the scope of this project we are interested in creating a dataset for deformable, semantic SLAM. The goal is to recognize, segment and reconstruct different object categories in real, indoor environments. The focus is live, dynamic objects, such as people and animals. Still, to this date, accurate segmentation in real-time of deformable surfaces is a challenging problem.
Visit our lab at Garching.
internal project page
Please contact Prof. Nassir Navab
, Dr. David Tan
, Johanna Wald
for available student projects within this research project.