Abstract

While deep learning approaches have demonstrated impressive results in a wide variety of computer vision and image processing tasks, the demand of large scale training data with annotations does hinder its steps to conquer several practical applications. In this talk I will introduce few of my recent works, in which I use deep learning models based on the unsupervised learning scenario to tackle several important computer vision problems and interesting applications of image processing. The talk will cover the topics of stereo matching, optical flow estimation, semantic segmentation, saliency-driven image manipulation, and the reverse/serial style transfer.

Bio

"Wei-Chen Chiu" (邱維辰) received the B.S. degree in Electrical Engineering and Computer Science and the M.S. degree in Computer Science from National Chiao Tung University (Hsinchu, Taiwan) in 2008 and 2009 respectively. He further received Doctor of Engineering Science (Dr.-Ing.) from Max Planck Institute for Informatics (Saarbrucken, Germany) in 2016. He joints Department of Computer Science, National Chiao Tung University as an assistant professor from August 2017 and leads the Enriched Vision Applications Laboratory. He was a postdoctoral researcher in Research Center for Information Technology Innovation, Academia Sinica, from Feb. to July. 2017, and a research scientist in a Taiwanese startup, Viscovery, from Aug. 2016 to Jan. 2017. His current research interests generally include computer vision, machine learning, and deep learning, with special focus on generative models.

All faculties and students are welcome to join.