Title: | CMIV Colloquium: Semi-Supervised Subspace Learning for Mumford-Shah Model Based Texture Segmentation |
Speaker: | Dr. Andy Yip, Department of Mathematics, National University of Singapore, Singapore |
Time/Place: | 14:30 - 15:30 DLB514, David C. Lam Building, Shaw Campus, Hong Kong Baptist University |
Abstract: | We propose a novel image segmentation model which incorporates subspace clustering techniques into a Mumford-Shah model to solve texture segmentation problems. While the natural unsupervised approach to learn a feature subspace can easily be trapped in a local solution, we propose a novel semi-supervised optimization algorithm that makes use of information derived from both the intermediate segmentation results and the regions-of-interest (ROI) selected by the user to determine the optimal subspaces of the target regions. These subspaces are embedded into a Mumford-Shah objective function so that each segment of the optimal partition is homogeneous in its own subspace. The method outperforms standard Mumford-Shah models since it can separate textures which are less separated in the full feature space. Experimental results are presented to confirm the usefulness of subspace clustering in texture segmentation. |
Title: | Efficient Algorithms for Total Variation Based Image Reconstruction and Applications in Partially Parallel MR Imaging |
Speaker: | Dr. Xiaojing Ye, Department of Mathematics, University of Florida, USA |
Time/Place: | 11:00 - 12:00 FSC1217, Fong Shu Chuen Library, HSH Campus, Hong Kong Baptist University |
Abstract: | In this talk I will discuss several fast numerical algorithms for image reconstruction with total variation (TV) regularization. TV based image reconstruction has proved to be very promising in magnetic resonance (MR) imaging. However the solution of TV based reconstruction encounters two main difficulties on the computational aspect of the emerging MR imaging application known as partially parallel imaging (PPI): the inversion matrix is large and severely ill-conditioned, and the objective is nonsmooth. We will talk about three algorithms that tackle the problem using variable splitting, Lagrangian methods and optimal step size selection technique. The numerical results show that these algorithms outperform the state-of-the-arts in terms of effectiveness and robustness. |
We organize conferences and workshops every year. Hope we can see you in future.
Learn MoreProf. M. Cheng, Dr. Y. S. Hon, Dr. K. F. Lam, Prof. L. Ling, Dr. T. Tong and Prof. L. Zhu have been awarded research grants by Hong Kong Research Grant Council (RGC) — congratulations!
Learn MoreFollow HKBU Math