Search

Gait recognition enables the non-contact identification of individuals from a distance based on their walking patterns and body shapes. For vision-based gait recognition, covariates (e.g., clothing, baggage and background) can negatively impact identification. As a result, many existing studies extract gait features from silhouettes or skeletal information obtained through preprocessing, rather than directly from RGB image sequences. In contrast to preprocessing which relies on the fitting accuracy of models trained on different tasks, disentangled representation learning (DRL) is drawing attention as a method for directly extracting gait features from RGB image sequences. However, DRL learns to extract features of the target attribute from the differences among multiple inputs with various attributes, which means its separation performance depends on the variation and amount of the training data. In this study, aiming to enhance the variation and quantity of each subject’s videos, we propose a novel data augmentation pipeline by feature swapping for RGB-based gait recognition. To expand the variety of training data, features of posture and covariates separated through DRL are paired with features extracted from different individuals, which enables the generation of images of subjects with new attributes. Dynamic gait features are extracted through temporal modeling from pose features of each frame, not only from real images but also from generated ones. The experiments demonstrate that the proposed pipeline increases both the quality of generated images and the identification accuracy. The proposed method also outperforms the RGB-based state-of-the-art method in most settings.

Gait recognition is a biometric identification method based on individual walking patterns. This modality is applied in a wide range of applications, such as criminal investigations and identification systems, since it can be performed at a long distance and requires no cooperation of interests. In general, cameras are used for gait recognition systems, and previous studies have utilized depth information captured by RGB-D cameras, such as Microsoft Kinect. In recent years, multi-layer LiDAR sensors, which can obtain range images of a target at a range of over 100 m in real time, have attracted significant attention in the field of autonomous mobile robots and self-driving vehicles. Compared with general cameras, LiDAR sensors have rarely been used for biometrics due to the low point cloud densities captured at long distances. In this study, we focus on improving the robustness of gait recognition using LiDAR sensors under confounding conditions, specifically addressing the challenges posed by viewing angles and measurement distances. First, our recognition model employs a two-scale spatial resolution to enhance immunity to varying point cloud densities. In addition, this method learns the gait features from two invariant viewpoints (i.e., left-side and back views) generated by estimating the walking direction. Furthermore, we propose a novel attention block that adaptively recalibrates channel-wise weights to fuse the features from the aforementioned resolutions and viewpoints. Comprehensive experiments conducted on our dataset demonstrate that our model outperforms existing methods, particularly in cross-view, cross-distance challenges, and practical scenarios.

Abstract This study proposes a new software platform, called ROS2-TMS, for an informationally structured environment. An informationally structured environment is vital for developing intelligent service robots by embedding various sensors in the environment to enhance the sensing capability and intelligence of robots. Thus far, we have been developing a software platform, named ROS-TMS, for an informationally structured environment, which connects various sensors and robots using ROS architecture. In recent years, ROS2, a next-generation version of ROS, has been released. ROS2 has many advantages, such as enhanced security, QoS control, and support for various platforms. ROS2-TMS, a new version of ROS-TMS, is developed not only by porting existing modules in ROS-TMS, such as the control system for a communication robot, but also by adding useful functions utilizing new features in ROS2. For instance, we added a voice user interface to control robots and various devices in the environment, such as lights or a bed. In addition, we implemented a new task scheduler that provides a cancelation function to stop running tasks and improve the security of the platform.

Abstract Autonomous robots are expected to replace dangerous, dirty, and demanding (3D) jobs. At a theme park, surveillance, cleaning, and guiding tasks can be regarded as 3D jobs. The present paper attempts to develop an autonomous tour guide robot system and co-experience system at a large theme park. For realizing such autonomous service robots used in our daily environment, localization is one of the most important and fundamental functions. A number of localization techniques, including simultaneous localization and mapping, have been proposed. Although a global navigation satellite system (GNSS) is most commonly used in outdoor environments, its accuracy is approximately 10 m, which is inadequate for navigation of an autonomous service robot. Therefore, a GNSS is usually used together with other localization techniques, such as map matching or camera-based localization. In the present study, we adopt the quasi-zenith satellite system (QZSS), which became available in and around Japan in November 2018, for the localization of an autonomous service robot. The QZSS provides high-accuracy position information using quasi-zenith satellites (QZSs) and has a localization error of less than 10 centimeters. In the present paper, we compare the positioning performance of the QZSS and the real-time kinematic GPS and verify the stability and the accuracy of the QZSS in an outdoor environment. In addition, we introduce a tour guide robot system using the QZSS and present the results of a guided tour experiment in a theme park. On the other hand, based on the tour guide system, we also develop a co-experience robot system in a theme park, which realizes the sharing of experiences using an immersive VR display and the 5th-generation mobile communication system (5G). The robot is equipped with a 360-degree video camera and real-time 4K video is transmitted to a remote operator using the large communication capacity of the 5G network. The experimental results at a theme park showed that the guided tour experiment was successful and that the co-experience system allowed sharing of the experience with high immersion by a remote operator.

Abstract Automatic analysis of our daily lives and activities through a first-person lifelog camera provides us with opportunities to improve our life rhythms or to support our limited visual memories. Notably, to express the visual experiences, the task of generating captions from first-person lifelog images has been actively studied in recent years. First-person images involve scenes approximating what users actually see; therein, the visual cues are not enough to express the user’s context since the images are limited by his/her intention. Our challenge is to generate lifelog captions using a meta-perspective called “fourth-person vision”. The “fourth-person vision” is a novel concept which complementary exploits the visual information from the first-, second-, and third-person perspectives. First, we assume human–robot symbiotic scenarios that provide a second-person perspective from the camera mounted on the robot and a third-person perspective from the camera fixed in the symbiotic room. To validate our approach in this scenario, we collect perspective-aware lifelog videos and corresponding caption annotations. Subsequently, we propose a multi-perspective image captioning model composed of an image-wise salient region encoder, an attention module that adaptively fuses the salient regions, and a caption decoder that generates scene descriptions. We demonstrate that our proposed model based on the fourth-person concept can greatly improve the captioning performance against single- and double-perspective models.

Abstract Spatial change detection is a fundamental technique for finding the differences between two or more pieces of geometrical information. This technique is critical in some robotic applications, such as search and rescue, security, and surveillance. In these applications, it is desirable to find the differences quickly and robustly. The present paper proposes a fast and robust spatial change detection technique for a mobile robot using an on-board range sensors and a highly precise 3D map created by a 3D laser scanner. This technique first converts point clouds in a map and measured data to grid data (ND voxels) using normal distributions transform. The voxels in the map and the measured data are then compared according to the features of the ND voxels. Three techniques are introduced to make the proposed system robust for noise, that is, classification of point distribution, overlapping of voxels, and voting using consecutive sensing. The present paper shows the results of indoor and outdoor experiments using an RGB-D camera and an omni-directional laser scanner mounted on a mobile robot to confirm the performance of the proposed technique.

Abstract This paper proposes a navigation system of a personal mobility robot in informationally structured and unstructured environments. First, we introduce a group of robots named Portable Go, which are equipped with laser range finders, gyros, and omni-directional wheels. The Portable Go robots expand the informationally structured environment by deploying in the informationally unstructured environment in advance. Then, a personal mobility robot based on an electric wheelchair is guided by eleven Portable Go robots in the new informationally structured environment which is just created by the Portable Go robots. Through navigation experiments, we verify that the proposed system navigates the personal mobility robot from the informationally structured environment to the informationally unstructured environment smoothly by using the Portable Go robots.

This paper introduces an enhanced MSM (Mutual Subspace Method) methodology for gait recognition, to provide robustness to variations in walking speed. The enhanced MSM (eMSM) methodology expands and adapts the MSM, commonly used for face recognition, which is a static/physiological biometric, to gait recognition, which is a dynamic/behavioral biometrics. To address the loss of accuracy during calculation of the covariance matrix in the PCA step of MSM, we use a 2D PCA-based mutual subspace. Furhtermore, to enhance the discrimination capability, we rotate images over a number of angles, which enables us to extract richer gait features to then be fused by a boosting method. The eMSM methodology is evaluated on existing data sets which provide variable walking speed, i.e. CASIA-C and OU-ISIR gait databases, and it is shown to outperform state-of-the art methods. While the enhancement to MSM discussed in this paper uses combinations of 2D-PCA, rotation, boosting, other combinations of operations may also be advantageous.

Abstract An informationally structured environment (ISE) is a key technology for realizing service robots in daily life in the near future. In ISE, the information of the service robot and its surroundings is structured and provided to the robot on time and on demand. We started the development of a management system for ISE named TMS (Town Management System) in Robot Town Project in 2005. Since then we are continuing our efforts for the improvement of the performance and the enhancement of the functions of TMS. In this paper, we propose the latest system of TMS named ROS-TMS 5.0, which adopts ROS to utilize high scalability and rich resources of ROS. Next, we introduce the hardware platform of ISE called Big Sensor Box which incorporates various sensors under ROS-TMS management and operates service robots based on structured information. Robot service experiments including watching service of a care receiver, voice control of a communication robot and a robotic bed, and ing information by voice are also conducted in Big Sensor Box.

This article presents the algorithm for real-time distant 3D tracking by RobotEye RE05, the 3D laser scanner using gimbal mirror mechanism. It is comprised of real-time extraction of moving objects by 3D background subtraction for 3D point cloud scanned by the scanner, tracking of a moving object, and scanning densely around it. For recognizing moving objects, high-speed background subtraction and clustering algorithms for point cloud data are developed. In the background subtraction, background data which contains no moving object is measured, and the distance map is generated, which contains distances of points in sections divided into every small azimuth and elevation angles. The map is adjusted to reduce noises caused by the measurement error and noises around object edges. Then, current data is measured which contains data of moving objects and the distance map is generated. If the distance in the map of current data is nearer than background data, the point is remained as the foreground data. Before clustering, edge noises in the foreground data are removed by the filter newly developed based on statistical outlier removal. This filter is improved by adjusting the number of neighbor points used for searching and mean distance with the neighbor points using the distance of each point. In the clustering, intervals of azimuth, elevation angles and distances of each point are considered to obtain proper results in any distances. Finally, the position and the width of moving objects are computed from the clusters. For following and scanning the moving object, the new function named Tracking Scan was developed, which is new scan motion of RobotEye RE05 for tracking application. This scanner can scan appointed range in azimuth 360deg and elevation ±35deg by desired density. Moreover, the range and density can be updated during scan in Tracking Scan. At last, the experiment result of moving person tracking is described.

The application of assistive technologies for elderly people is one of the most promising and interesting scenarios for intelligent technologies in the present and near future. Moreover, the improvement of the quality of life for the elderly is one of the first priorities in modern countries and societies. In this work, we present an informationally structured room that is aimed at supporting the daily life activities of elderly people. This room integrates different sensor modalities in a natural and non-invasive way inside the environment. The information gathered by the sensors is processed and sent to a centralized management system, which makes it available to a service robot assisting the people. One important restriction of our intelligent room is reducing as much as possible any interference with daily activities. Finally, this paper presents several experiments and situations using our intelligent environment in cooperation with our service robot.

This paper describes a new method of measuring the position of everyday objects and a robot on the floor using distance and reflectance acquired by laser range finder (LRF). The information obtained by this method is important for a service robot working in a human daily life environment. Our method uses only one LRF together with a mirror installed on the wall. Moreover, since the area of sensing is limited to a LRF scanning plane parallel to the floor and just a few centimeters above the floor, the scanning covers the whole room with minimal invasion of privacy of a resident, and occlusion problem is mitigated by using mirror. We use the reflection intensity and position information obtained from the target surface. Although it is not possible to identify all objects by additionally using reflection values, it would be easier to identify unknown objects if we can eliminate easily identifiable objects by reflectance. In addition, we propose a method for measuring the robot’s pose using the tag which has the encoded reflection pattern optically identified by the LRF. Our experimental results validate the effectiveness of the proposed method.

The identification of a person from gait images is generally sensitive to appearance changes, such as variations of clothes and belongings. One possibility to deal with this problem is to collect possible subjects’ appearance changes in a database. However, it is almost impossible to predict all appearance changes in advance. In this paper, we propose a novel method, which allows robustly identifying people in spite of changes in appearance, without using a database of predicted appearance changes. In the proposed method, firstly, the human body image is divided into multiple areas, and features for each area are extracted. Next, a matching weight for each area is estimated based on the similarity between the extracted features and those in the database for standard clothes. Finally, the subject is identified by weighted integration of similarities in all areas. Experiments using the gait database CASIA show the best correct classification rate compared with conventional methods experiments.

The categorization of places in indoor environments is an important capability for service robots working and interacting with humans. In this paper we present a method to categorize different areas in indoor environments using a mobile robot equipped with a Kinect camera. Our approach transforms depth and grey scale images taken at each place into histograms of local binary patterns (LBPs) whose dimensionality is further reduced following a uniform criterion. The histograms are then combined into a single feature vector which is categorized using a supervised method. In this work we compare the performance of support vector machines and random forests as supervised classifiers. Finally, we apply our technique to distinguish five different place categories: corridors, laboratories, offices, kitchens, and study rooms. Experimental results show that we can categorize these places with high accuracy using our approach.