Files in this item



application/pdfECE499-Sp2015-lowney.pdf (1MB)
(no description provided)PDF


Title:Camera Pose Estimation of RGB-D Sensors Using Particle Filtering
Author(s):Lowney, Michael
Contributor(s):Do, Minh N.
Subject(s):Particle filter
Camera pose
Abstract:The number of commercially available cameras that measure both color and depth (RGB-D) has increased significantly in the past ten years. In this paper we explore how particle filtering can be used to estimate the position and orientation of an RGB-D camera. A particle filter is a nonparametric Bayes filter which uses samples of the state space to approximate a posterior. Localization plays a key role in many robotics and augmented reality applications. We compare the particle filter to other Bayesian techniques and make a case for the use of particle filters. Our methods include running simulations in MATLAB to better understand the particle filter and its tradeoffs. By using depth data from the camera and comparing it to a map of the environment we are able to estimate the position of the camera. In our work we empirically examine how certain factors such as the number of particles, and the quality of the map used affect the pose estimation.
Issue Date:2015-05
Date Available in IDEALS:2015-08-03

This item appears in the following Collection(s)

Item Statistics