Robust motion segmentation with subspace constraints
Motion segmentation is an important task in computer vision with many applications such as dynamic scene understanding and multi-body structure from motion. When the point correspondences across frames are given, motion segmentation can be addressed as a subspace clustering problem under an affine camera model. In the first two parts of this thesis, we target the general subspace clustering problem and propose two novel methods, namely Efficient Dense Subspace...[Show more]
|Collections||Open Access Theses|
|Ji_Thesis_2016.pdf||6.82 MB||Adobe PDF|
Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.