• The Journal of Korea Robotics Society
• /
• v.14 no.2
• /
• pp.114-121
• /
• 2019

This paper presents a new benchmark system for visual odometry (VO) and monocular depth estimation (MDE). As deep learning has become a key technology in computer vision, many researchers are trying to apply deep learning to VO and MDE. Just a couple of years ago, they were independently studied in a supervised way, but now they are coupled and trained together in an unsupervised way. However, before designing fancy models and losses, we have to customize datasets to use them for training and testing. After training, the model has to be compared with the existing models, which is also a huge burden. The benchmark provides input dataset ready-to-use for VO and MDE research in 'tfrecords' format and output dataset that includes model checkpoints and inference results of the existing models. It also provides various tools for data formatting, training, and evaluation. In the experiments, the exsiting models were evaluated to verify their performances presented in the corresponding papers and we found that the evaluation result is inferior to the presented performances.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.3
• /
• pp.160-173
• /
• 2013

In this paper, we propose a new visual odometry technique by combining a forward-looking stereo camera and a backward-looking monocular camera. The main goal of the proposed technique is to identify the location of a moving vehicle which travels long distance and comes back to the initial position in urban road environments. While the vehicle is moving to the destination, a global 3D map is updated continuously by a stereo visual odometry technique using a graph theorem. Once the vehicle reaches the destination and begins to come back to the initial position, a map-based monocular visual odometry technqieu is used. To estimate the position of the returning vehicle accurately, 2D features in the backward-looking camera image and the global map are matched. In addition, we utilize the previous matching nodes to limit the search ranges of the next vehicle position in the global map. Through two navigation paths, we analyze the accuracy of the proposed method.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.4
• /
• pp.216-223
• /
• 2015

The autonomous underwater vehicle (AUV) is being widely researched in order to achieve superior performance when working in hazardous environments. This research focuses on using image processing techniques to estimate the AUV's egomotion and the changes in orientation, based on image frames from different time frames captured from a single high-definition web camera attached to the bottom of the AUV. A visual odometry application is integrated with other sensors. An internal measurement unit (IMU) sensor is used to determine a correct set of answers corresponding to a homography motion equation. A pressure sensor is used to resolve image scale ambiguity. Uncertainty estimation is computed to correct drift that occurs in the system by using a Jacobian method, singular value decomposition, and backward and forward error propagation.

• The Journal of Korea Robotics Society
• /
• v.3 no.4
• /
• pp.263-269
• /
• 2008

Visual odometry is a popular approach to estimating robot motion using a monocular or stereo camera. This paper proposes a novel visual odometry scheme using a stereo camera for robust estimation of a 6 DOF motion in the dynamic environment. The false results of feature matching and the uncertainty of depth information provided by the camera can generate the outliers which deteriorate the estimation. The outliers are removed by analyzing the magnitude histogram of the motion vector of the corresponding features and the RANSAC algorithm. The features extracted from a dynamic object such as a human also makes the motion estimation inaccurate. To eliminate the effect of a dynamic object, several candidates of dynamic objects are generated by clustering the 3D position of features and each candidate is checked based on the standard deviation of features on whether it is a real dynamic object or not. The accuracy and practicality of the proposed scheme are verified by several experiments and comparisons with both IMU and wheel-based odometry. It is shown that the proposed scheme works well when wheel slip occurs or dynamic objects exist.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.1095-1109
• /
• 2021

The technology used to recognize the location and surroundings of autonomous vehicles is called SLAM. SLAM standsfor Simultaneously Localization and Mapping and hasrecently been actively utilized in research on autonomous vehicles,starting with robotic research. Expensive GPS, INS, LiDAR, RADAR, and Wheel Odometry allow precise magnetic positioning and mapping in centimeters. However, if it can secure similar accuracy as using cheaper Cameras and GPS data, it will contribute to advancing the era of autonomous driving. In this paper, we present a method for converging monocular camera with RTK-enabled GPS data to perform RMSE 33.7 cm localization and mapping on the urban road.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.2
• /
• pp.164-170
• /
• 2011

This paper proposes a novel monocular vision-based SLAM (Simultaneous Localization and Mapping) method using both position and orientation information of ceiling lamps. Conventional approaches used corner or line features as landmarks in their SLAM algorithms, but these methods were often unable to achieve stable navigation due to a lack of reliable visual features on the ceiling. Since lamp features are usually placed some distances from each other in indoor environments, they can be robustly detected and used as reliable landmarks. We used both the position and orientation of a lamp feature to accurately estimate the robot pose. Its orientation is obtained by calculating the principal axis from the pixel distribution of the lamp area. Both corner and lamp features are used as landmarks in the EKF (Extended Kalman Filter) to increase the stability of the SLAM process. Experimental results show that the proposed scheme works successfully in various indoor environments.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.6
• /
• pp.99-104
• /
• 2021

The technology of estimating a movement trajectory using a monocular camera such as a smartphone and composing a surrounding 3D image is key not only in indoor positioning but also in the metaverse service. The most important thing in this technique is to estimate the coordinates of the moving camera center. In this paper, a new algorithm for geometrically estimating the moving distance is proposed. The coordinates of the 3D object point are obtained from the first and second photos, and the movement distance vector is obtained using the matching feature points of the first and third photos. Then, while moving the coordinates of the origin of the third camera, a position where the 3D object point and the feature point of the third picture coincide is obtained. Its possibility and accuracy were verified by applying it to actual continuous image data.

• Journal of Positioning, Navigation, and Timing
• /
• v.12 no.2
• /
• pp.129-140
• /
• 2023

Autonomous driving systems are likely to be operated in various complex environments. However, the well-known integrated Global Navigation Satellite System (GNSS)/Inertial Navigation System (INS), which is currently the major source for absolute position information, still has difficulties in accurate positioning in harsh signal environments such as urban canyons. To overcome these difficulties, integrated Visual/Inertial/GNSS (VIG) navigation systems have been extensively studied in various areas. Recently, a Compressed-State Constraint Kalman Filter (CSCKF)-based VIG navigation system (CSCKF-VIG) using a monocular camera, an Inertial Measurement Unit (IMU), and GNSS receivers has been studied with the aim of providing robust and accurate position information in urban areas. For this new filter-based navigation system, on the basis of time-propagation measurement fusion theory, unnecessary camera states are not required in the system state. This paper presents a performance evaluation of the CSCKF-VIG system compared to other conventional navigation systems. First, the CSCKF-VIG is introduced in detail compared to the well-known Multi-State Constraint Kalman Filter (MSCKF). The CSCKF-VIG system is then evaluated by a field experiment in different GNSS availability situations. The results show that accuracy is improved in the GNSS-degraded environment compared to that of the conventional systems.