Zhaopeng Cui - PhD Depth Examination

PhD Depth Examination: Camera Pose Estimation in Structure-From-Motion - Zhaopeng Cui


Structure-from-motion (SfM) is a fundamental problem in computer vision. It refers to the process of estimating 3d scene structures and camera poses simultaneously from multiple 2d images. Conventional SfM systems often consist of three steps:

1) estimation of point correspondences and relative poses between images

2) estimation of camera motions and 3d points, and

3) nonlinear optimization for both camera motions and 3d points.

While there are well established theories for the first and the third steps, the second step in existing systems are often ad-hoc and heuristic. In this report, we start with introducing the SfM pipeline. Then we review the incremental camera motion estimation methods which add cameras one by one to the global coordinate system. Thirdly, we review the recent global camera motion estimation methods which try to recover all camera poses together. At last, we will provide some insights for future work based on the analysis of previous methods

Ph.D. Examining Committee:

Dr. Ping Tan, Senior Supervisor

Dr. Greg Mori, Supervisor

Dr. Mark Drew, Examiner


Friday April 8, 2016 9:00 a.m. TASC1 9204 West
Friday, April 8, 2016 - 09:00 to 11:00