scholarly journals Real-world Comparison of Visual Odometry Methods

2020 ◽  
Author(s):  
Alexandre Alapetite ◽  
Zhongyu Wang ◽  
John Paulin Hansen ◽  
Marcin Zajączkowski ◽  
Mikolaj Patalan
Keyword(s):  
Ground Truth ◽  
Visual Odometry ◽  
Visual Features ◽  
Fully Integrated ◽  
Result Show ◽  
Comparable Performance ◽  
Internal Information ◽  
Robot Position ◽  
Visual Clues ◽  
Do So

Positioning is an essential aspect of robot navigation, and visual odometry an important technique for continuous updating the internal information about robot position, especially indoors without GPS. Visual odometry is using one or more cameras to find visual clues and estimate robot movements in 3D relatively. Recent progress has been made, especially with fully integrated systems such as the RealSense T265 from Intel, which is the focus of this article. We compare between each other three visual odometry systems and one wheel odometry, on a ground robot. We do so in 8 scenarios, varying the speed, the number of visual features, and with or without humans walking in the field of view. We continuously measure the position error in translation and rotation thanks to a ground truth positioning system. Our result show that all odometry systems are challenged, but in different ways. In average, ORB-SLAM2 has the poorer results, while the RealSense T265 and the Zed Mini have comparable performance. In conclusion, a single odometry system might still not be sufficient, so using multiple instances and sensor fusion approaches are necessary while waiting for additional research and further improved products.

scholarly journals Comparison of Three Off-the-Shelf Visual Odometry Systems

Robotics ◽  
2020 ◽  
Vol 9 (3) ◽  
pp. 56 ◽  
Author(s):  
Alexandre Alapetite ◽  
Zhongyu Wang ◽  
John Paulin Hansen ◽  
Marcin Zajączkowski ◽  
Mikołaj Patalan
Keyword(s):  
Ground Truth ◽  
Visual Odometry ◽  
Measurement Unit ◽  
Positioning System ◽  
Fully Integrated ◽  
Global Positioning ◽  
Comparable Performance ◽  
Robot Position ◽  
Visual Clues ◽  
Better Than

Positioning is an essential aspect of robot navigation, and visual odometry an important technique for continuous updating the internal information about robot position, especially indoors without GPS (Global Positioning System). Visual odometry is using one or more cameras to find visual clues and estimate robot movements in 3D relatively. Recent progress has been made, especially with fully integrated systems such as the RealSense T265 from Intel, which is the focus of this article. We compare between each other three visual odometry systems (and one wheel odometry, as a known baseline), on a ground robot. We do so in eight scenarios, varying the speed, the number of visual features, and with or without humans walking in the field of view. We continuously measure the position error in translation and rotation thanks to a ground truth positioning system. Our result shows that all odometry systems are challenged, but in different ways. The RealSense T265 and the ZED Mini have comparable performance, better than our baseline ORB-SLAM2 (mono-lens without inertial measurement unit (IMU)) but not excellent. In conclusion, a single odometry system might still not be sufficient, so using multiple instances and sensor fusion approaches are necessary while waiting for additional research and further improved products.

scholarly journals Static and Dynamic Evaluation of an UWB Localization System for Industrial Applications

Sci ◽  
2020 ◽  
Vol 2 (1) ◽  
pp. 7 ◽  
Author(s):  
Mickaël Delamare ◽  
Remi Boutteau ◽  
Xavier Savatier ◽  
Nicolas Iriart
Keyword(s):  
Motion Capture ◽  
Indoor Environment ◽  
Indoor Localization ◽  
Ground Truth ◽  
Industrial Applications ◽  
Ultra Wideband ◽  
Complex Environments ◽  
Experimental Protocol ◽  
Localization Systems ◽  
Do So

Many applications in the context of Industry 4.0 require precise localization. However, indoor localization remains an open problem, especially in complex environments such as industrial environments. In recent years, we have seen the emergence of Ultra WideBand (UWB) localization systems. The aim of this article is to evaluate the performance of a UWB system to estimate the position of a person moving in an indoor environment. To do so, we implemented an experimental protocol to evaluate the accuracy of the UWB system both statically and dynamically. The UWB system is compared to a ground truth obtained by a motion capture system with a millimetric accuracy.

scholarly journals Static and Dynamic Evaluation of an UWB Localization System for Industrial Applications

Sci ◽  
2019 ◽  
Vol 1 (3) ◽  
pp. 62 ◽  
Author(s):  
Mickael Delamare ◽  
Remi Boutteau ◽  
Xavier Savatier ◽  
Nicolas Iriart
Keyword(s):  
Motion Capture ◽  
Indoor Environment ◽  
Indoor Localization ◽  
Ground Truth ◽  
Industrial Applications ◽  
Ultra Wideband ◽  
Complex Environments ◽  
Experimental Protocol ◽  
Localization Systems ◽  
Do So

Many applications in the context of Industry 4.0 require precise localization. However, indoor localization remains an open problem, especially in complex environments such as industrial environments. In recent years, we have seen the emergence of Ultra WideBand (UWB) localization systems. The aim of this article is to evaluate the performance of a UWB system to estimate the position of a person moving in an indoor environment. To do so, we implemented an experimental protocol to evaluate the accuracy of the UWB system both statically and dynamically. The UWB system is compared to a ground truth obtained by a motion capture system with a millimetric accuracy.

scholarly journals Monocular SLAM for Visual Odometry: A Full Approach to the Delayed Inverse-Depth Feature Initialization Method

2012 ◽  
Vol 2012 ◽  
pp. 1-26 ◽  
Author(s):  
Rodrigo Munguía ◽  
Antoni Grau
Keyword(s):  
Sensory Input ◽  
A Priori ◽  
Real Data ◽  
Visual Odometry ◽  
Visual Features ◽  
Single Camera ◽  
Localization And Mapping ◽  
Monocular Slam ◽  
Freely Moving ◽  
Depth Feature

This paper describes in a detailed manner a method to implement a simultaneous localization and mapping (SLAM) system based on monocular vision for applications of visual odometry, appearance-based sensing, and emulation of range-bearing measurements. SLAM techniques are required to operate mobile robots ina prioriunknown environments using only on-board sensors to simultaneously build a map of their surroundings; this map will be needed for the robot to track its position. In this context, the 6-DOF (degree of freedom) monocular camera case (monocular SLAM) possibly represents the harder variant of SLAM. In monocular SLAM, a single camera, which is freely moving through its environment, represents the sole sensory input to the system. The method proposed in this paper is based on a technique called delayed inverse-depth feature initialization, which is intended to initialize new visual features on the system. In this work, detailed formulation, extended discussions, and experiments with real data are presented in order to validate and to show the performance of the proposal.

scholarly journals Unsupervised Deep Learning-Based RGB-D Visual Odometry

2020 ◽  
Vol 10 (16) ◽  
pp. 5426 ◽  
Author(s):  
Qiang Liu ◽  
Haidong Zhang ◽  
Yiming Xu ◽  
Li Wang
Keyword(s):  
Deep Learning ◽  
Feature Matching ◽  
Ground Truth ◽  
Visual Odometry ◽  
Depth Images ◽  
Network Training ◽  
Stream Structure ◽  
Unsupervised Deep Learning ◽  
Rgb Images ◽  
Learning Frameworks

Recently, deep learning frameworks have been deployed in visual odometry systems and achieved comparable results to traditional feature matching based systems. However, most deep learning-based frameworks inevitably need labeled data as ground truth for training. On the other hand, monocular odometry systems are incapable of restoring absolute scale. External or prior information has to be introduced for scale recovery. To solve these problems, we present a novel deep learning-based RGB-D visual odometry system. Our two main contributions are: (i) during network training and pose estimation, the depth images are fed into the network to form a dual-stream structure with the RGB images, and a dual-stream deep neural network is proposed. (ii) the system adopts an unsupervised end-to-end training method, thus the labor-intensive data labeling task is not required. We have tested our system on the KITTI dataset, and results show that the proposed RGB-D Visual Odometry (VO) system has obvious advantages over other state-of-the-art systems in terms of both translation and rotation errors.

scholarly journals Exploiting Attitude Sensing in Vision-Based Navigation for an Airship

2009 ◽  
Vol 2009 ◽  
pp. 1-16 ◽  
Author(s):  
Luiz G. B. Mirisola ◽  
Jorge Dias
Keyword(s):  
Ground Plane ◽  
Translational Motion ◽  
Ground Truth ◽  
Visual Odometry ◽  
Magnetic Sensors ◽  
Controlled Environment ◽  
Translation Model ◽  
Graph Optimization ◽  
Vision Based Navigation ◽  
Previous Image

An Attitude Heading Reference System (AHRS) is used to compensate for rotational motion, facilitating vision-based navigation above smooth terrain by generating virtual images to simulate pure translation movement. The AHRS combines inertial and earth field magnetic sensors to provide absolute orientation measurements, and our recently developed calibration routine determines the rotation between the frames of reference of the AHRS and the monocular camera. In this way, the rotation is compensated, and the remaining translational motion is recovered by directly finding a rigid transformation to register corresponding scene coordinates. With a horizontal ground plane, the pure translation model performs more accurately than image-only approaches, and this is evidenced by recovering the trajectory of our airship UAV and comparing with GPS data. Visual odometry is also fused with the GPS, and ground plane maps are generated from the estimated vehicle poses and used to evaluate the results. Finally, loop closure is detected by looking for a previous image of the same area, and an open source SLAM package based in 3D graph optimization is employed to correct the visual odometry drift. The accuracy of the height estimation is also evaluated against ground truth in a controlled environment.

scholarly journals CALIBRATION AND VALIDATION OF THE INTEL T265 FOR VISUAL LOCALISATION AND TRACKING UNDERWATER

2021 ◽  
Vol XLIII-B2-2021 ◽  
pp. 635-641
Author(s):  
T. Appelt ◽  
J. van der Lucht ◽  
M. Bleier ◽  
A. Nüchter
Keyword(s):  
Autonomous Underwater Vehicle ◽  
Low Cost ◽  
Ground Truth ◽  
Visual Odometry ◽  
Chaotic Motions ◽  
Stereo Camera ◽  
Accuracy And Precision ◽  
Complex Solutions ◽  
Future Work ◽  
Distortion Parameters

Abstract. Localization and navigation for autonomous underwater vehicle (AUV) has always been a major challenge and many situations complex solutions had to be devised. One of the main approaches is visual odometry using a stereo camera. In this study, the Intel T265 fisheye stereo camera has been calibrated and tested to determine it’s usability for localisation and navigation under water as an alternative to more complex systems. Firstly the Intel T265 fisheye stereo camera was appropriately calibrated inside a water filled container. This calibration consisting of camera and distortion parameters got programmed onto the T265 fisheye stereo camera to take the differences between land and underwater usage into account. Successive the calibration, the accuracy and the precision of the T265 fisheye stereo camera were tested using a linear, a circular and finally a chaotic motion. This includes a review of the localisation and tracking of the cameras visual odometry compared to a ground truth provided by an OptiTrack V120:Trio to account for scaling, accuracy and precision. Also experiments to determine the usability with fast chaotic motions were performed and analysed. Finally, a conclusion concerning the applicability of the Intel T265 fisheye stereo camera, the challenges using this model, the possibilities for low cost operations and the main challenges for future work is conducted.

A SIMILARITY METRIC IN IMAGE SEARCHING

2007 ◽  
Vol 07 (02) ◽  
pp. 211-225
Author(s):  
XUELONG LI ◽  
JING LI ◽  
DACHENG TAO ◽  
YUAN YUAN
Keyword(s):  
Signal Processing ◽  
System Performance ◽  
Ground Truth ◽  
Point Of View ◽  
Visual Features ◽  
Similarity Metrics ◽  
Graphical Interface ◽  
Similarity Metric ◽  
Processing Point ◽  
Small Image

Similarity metric is a key component in query-by-example image searching with visual features. After extraction of image visual features, the scheme of computing their similarities can affect the system performance dramatically — the image searching results are normally displayed in decreasing order of similarity (alternatively, increasing order of distance) on the graphical interface for end users. Unfortunately, conventional similarity metrics, in image searching with visual features, usually encounter several difficulties, namely, lighting, background, and viewpoint problems. From the signal processing point of view, this paper introduces a novel similarity metric and therefore reduces the above three problems to some extent. The effectiveness of this newly developed similarity metric is demonstrated by a set of experiments upon a small image ground truth.

Sign in / Sign up

Export Citation Format

Share Document