题目： 偏微分方程与优化在图像处理中的应用(I) (II)(III)

报告人：刘万泉（澳大利亚科廷大学）

时间：  2020/10/20  15:00-17:00

地点：腾讯会议 ID：105 204 662  会议密码：123456

报告摘要：

In this talk, we will talk about the image understanding problem based on PDE and optimization techniques. I first talk about the effectiveness of variational and PDE based methods for illusory contour reconstruction and image segmentation, and then design the corresponding optimization algorithms for efficiency improvement. Next, I talk about the  variational image segmentation problems in the optimization framework of stochastic programming, tackling diverse segmentation problems with random noises. Finallly, I focus on exploring the fusion approaches integrating varaitional models and deep neural networks for challenging image tasks using unsupervised and supervised learning respectively.

报告人简介:

Dr. Wanquan Liu received the BSc degree in Applied Mathematics from Qufu Normal University, P. R. China, in 1985, the MSc degree in Control Theory and Operation Research from Chinese Academy of Science in 1988, and the PhD degree in Electrical Engineering from Shanghai Jiaotong University, in 1993. He once held the ARC Fellowship, U2000 Fellowship and JSPS Fellowship and attracted research funds from different resources over 2.7 million dollars. He is currently an Associate Professor in the Department of Computing at Curtin University and is in editorial board for seven international journals. His current research interests include large-scale pattern recognition, signal processing, machine learning, and computer vision. He is the editor-in-chief for Journal of Mathematical Foundation of Computing (MFC).