Vision-Only Localization

Abstract

Abstract-Autonomous and intelligent vehicles will undoubtedly depend on an accurate ego localization solution. Global navigation satellite systems (GNSS) suffer from multipath propagation rendering this solution insufficient. Herein we present a real time system for six degrees of freedom (DOF) ego localization that uses only a single monocular camera. The camera image is harnessed to yield an ego pose relative to a previously computed visual map. We describe a process to automatically extract the ingredients of this map from stereoscopic image sequences. These include a mapping trajectory relative to the first pose, global scene signatures and local landmark descriptors. The localization algorithm then consists of a topological localization step that completely obviates the need for any global positioning sensors like GNSS. A metric refinement step that recovers an accurate metric pose is subsequently applied. Metric localization recovers the ego pose in a factor graph optimization process based on local landmarks. We demonstrate a centimeter level accuracy by a set of experiments in an urban environment. To this end, two localization estimates are computed for two independent cameras mounted on the same vehicle. These two independent trajectories are thereafter compared for consistency. Finally, we present qualitative experiments of an augmented reality (AR) system that depends on the aforementioned localization solution. Several screen shots of the AR system are shown confirming centimeter level accuracy and sub degree angular precision.

Image-Based Localization for Indoor Environment Using Mobile Phone

“…Therefore, the query image can be misclassified in some case. There are also methods using stereo images [1,3,7]. The stereo images can provide depth information which is helpful for 3D reconstruction.…”

Path tracking by a mobile robot equipped with only a downward facing camera

“…For motion estimation, there are several advantages of using visual odometry to track feature points in images of scenes from a camera mounted on the robot [4][5][6][7]. One advantage is that it is applicable to mobile robots with various types of traveling mechanism because its non-contact sensing method precludes mechanistic problems including wheel slippage and skids [8,9].…”

Position Interpolation Using Feature Point Scale for Decimeter Visual Localization

“…Visual localization methods typically use a pre-built database of image descriptors or features captured by a vehicle-mounted camera. Localization can either be performed using extracted feature points [11], [12], [23], or by matching query images to database images using whole image similarity [2], [6], [22], [24]. The correct database image match can be determined either using Dynamic Time Warping (DTW) with an image similarity measure [11], [22], [23] or a Bayes filter [2], [6], [24].…”

“…Methods using feature points have the advantage of robustness to occlusions and lane changes, but also come with increased complexity. Most techniques employing feature points calculate the camera pose relative to the matched database frame [10], [12]. This requires calculation of the essential matrix, usually together with iterative processes such as Random Sample Consensus (RANSAC) [8] and bundle adjustment.…”