ToF camera ego-motion estimation

Abstract

In this paper, three approaches for ego-motion estimation using Time-of-Flight (ToF) camera data are evaluated. Ego-motion is defined as a process of estimating a camera's pose relative to some initial pose using the camera's image sequence. The ToF camera is characterised with a number error models. These models are used to design several filters that are applied on point cloud data. Iterative Closest Point (ICP) is applied on the consecutive range images of the ToF camera to estimate relative pose transform which is used for egomotion estimation. We implemented two variants of ICP namely point-to-point and point-to-plane. A feature based ego-motion approach that detects and tracks features on the amplitude images and use their corresponding 3D points to estimate the relative transformation is implemented. These approaches are evaluated using the groundtruth provided by the vicon system.