TITLE: Point Cloud Registration Using Quantile Assignment
by Ecenur Oğuz, M.S. in Industrial Engineering
Advisor: Prof. Mustafa Ç. Pınar and Prof. Oya Karaşan
Date & Time: July 12, 2023 Wednesday at 10:30
This is an online seminar. To obtain event details please send a message to department.
ABSTRACT:
Point cloud registration is a fundamental problem in computer vision with a wide range of applications. The problem mainly consists of three parts: feature estimation, correspondence matching, and transformation estimation. We introduced the Quantile Assignment problem and proposed a solution algorithm to be used in a point cloud registration framework for establishing the correspondence set between the source and the target point clouds. We analyzed different common feature descriptors and transformation estimation methods to combine with our Quantile Assignment algorithm. The performance of these approaches together with our algorithm are tested with controlled experiments on a dataset we constructed using well-known 3D models. We detected the most suitable methods to combine with our approach and proposed a new end-to-end pairwise point cloud registration framework. Finally, we tested our framework on both indoor and outdoor benchmark datasets and compared our results with state-of-the-art point cloud registration methods in the literature.
BIO
Ecenur Oğuz received her B.S from the Department of Industrial Engineering Bilkent University in June 2021. She is currently pursuing M.S. degree in the Department of Industrial Engineering and working with Prof. Mustafa Ç. Pınar and Prof. Oya Karaşan on a point cloud registration algorithm. Alongside, her other research interests include combinatorial optimization.