Visit ShanghaiTech University | 中文 | How to find us
HOME > News and Events > Events
Beyond Structure From Motion: Exploring Structure Regularities for 3D Vision
Date: 2016/7/28             Browse: 670

Beyond Structure From Motion: Exploring Structure Regularities for 3D Vision

Speaker: Zihan Zhou

Time: Jul 28, 3:00pm - 4:00pm.

Location: Room 410, Teaching Building


Although significant progress has been made about the imaging devices and 3D reconstruction techniques over the past few decades, obtaining high-quality 3D models from images still remains a challenging problem in computer vision nowadays. Moreover, when directly utilized in real-world applications, the reconstructed models often lead to unsatisfactory or unstable results.

In this talk, I argue that much of the difficulties faced by the current techniques and applications can be alleviated by harnessing the structural regularities in the scene, such as parallel lines, planes, repetitive patterns, and all types of symmetries. Specifically, I demonstrate how structural regularities can be used to (1) develop novel algorithm for effective and efficient structure and motion recovery, and (2) facilitate a variety of 3D vision applications even in the absence of accurate and complete 3D models. As an ongoing research effort, I will also discuss new opportunities and challenges in exploiting structural regularities with large-scale visual data for emerging commercial and customer applications such as virtual reality, robotics, and autonomous driving.


Dr. Zihan Zhou is currently a research associate in the College of Information Sciences and Technology at the Pennsylvania State University. Before joining Penn State, he completed his Ph.D in Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign in 2013. In Summer 2016, he is a visiting professor in the School of Information Science and Technology, ShanghaiTech University. His research interest lies in the broad area of computer vision, image processing, and machine learning, with a focus in developing and analyzing new computational tools for efficient and robust discovery of low-dimensional data structure from high-dimensional observations. Recently, he is particularly interested in exploiting scene structure regularities for 3D reconstruction and its applications. To learn more, please visit his homepage:

SIST-Seminar 16058