TIMA SLAM: Tracking Independently and Mapping Altogether for an Uncalibrated Multi-Camera System.

Omer Faruk Ince,Jun-Sik Kim

doi:10.3390/s21020409

Abstract

We present a novel simultaneous localization and mapping (SLAM) system that extends the state-of-the-art ORB-SLAM2 for multi-camera usage without precalibration. In this system, each camera is tracked independently on a shared map, and the extrinsic parameters of each camera in the fixed multi-camera system are estimated online up to a scalar ambiguity (for RGB cameras). Thus, the laborious precalibration of extrinsic parameters between cameras becomes needless. By optimizing the map, the keyframe poses, and the relative poses of the multi-camera system simultaneously, observations from the multiple cameras are utilized robustly, and the accuracy of the shared map is improved. The system is not only compatible with RGB sensors but also works on RGB-D cameras. For RGB cameras, the performance of the system tested on the well-known EuRoC/ASL and KITTI datasets that are in the stereo configuration for indoor and outdoor environments, respectively, as well as our dataset that consists of three cameras with small overlapping regions. For the RGB-D tests, we created a dataset that consists of two cameras for an indoor environment. The experimental results showed that the proposed method successfully provides an accurate multi-camera SLAM system without precalibration of the multi-cameras.

Highlights

In robotics, vision-based simultaneous localization and mapping (SLAM) is a geometric problem of mapping an unknown environment while tracking the camera pose simultaneously
The third dataset (KIST) has been created by us that consists of sequences RGB and RGB-D
The RGB camera rig consists of three cameras with certain overlaps of their fields of view, and the RGB-D camera rig consists of two cameras that are set similar to stereo configuration

Summary

Introduction

Vision-based simultaneous localization and mapping (SLAM) is a geometric problem of mapping an unknown environment while tracking the camera pose simultaneously. Liu et al [1] show that a camera with a wide field of view provides more accurate camera localization and more robust mapping. In [2], it is explained the advantages of a wide field of view in the problem of place recognition and geometric SLAM. One of the solutions to obtain a wide field of view is to use a single camera with an omnidirectional lens. In [3], it shows that images taken with an omnidirectional lens suffer from wide image scale variation and low angular resolution. A pixel measurement of an omnidirectional camera is less accurate than that of a perspective camera

Objectives

Results

Conclusion