研究者詳細 - 米澤　拓郎

論文 - 米澤　拓郎

分割表示 301 件中 81 - 100 件目／全件表示 >>

XR Communication System for Remote Control Wheelchairs 査読有り

Shimosato, H; Hayashida, N; Urano, K; Yonezawa, T; Kawaguchi, N

PROCEEDINGS OF THE 12TH INTERNATIONAL CONFERENCE ON THE INTERNET OF THINGS 2022, IOT 2022 頁： 219 - 223 2022年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：ACM International Conference Proceeding Series

With the introduction of the IoT, communication is becoming more diverse. While IoT is becoming more and more widespread, there are some forms of human communication are being lost. The development of self-propelled robots has progressed, and electric and automated wheelchairs are also being utilized. A caregiver was originally required to operate a wheelchair. However, with the advent of IoT wheelchairs, wheelchairs can be operated by wheelchair users only, increasing convenience. On the other hand, wheelchairs have been used as a kind of communication between caregivers and wheelchair users, and the development of IoT wheelchairs has reduced the opportunities for communication between caregivers and wheelchair users. In this paper, we propose a method to reproduce communication scenarios in the following two scenarios using an IoT wheelchair and present the design and implementation of the prototype created.

DOI： 10.1145/3567445.3571111

その他リンク： https://dblp.uni-trier.de/db/conf/iot/iot2022.html#ShimosatoHUYK22

Synthetic People Flow: Privacy-Preserving Mobility Modeling from Large-Scale Location Data in Urban Areas 査読有り

Tamura, N; Urano, K; Aoki, S; Yonezawa, T; Kawaguchi, N

MOBILE AND UBIQUITOUS SYSTEMS: COMPUTING, NETWORKING AND SERVICES 419 巻頁： 553 - 567 2022年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：Lecture Notes of the Institute for Computer Sciences Social Informatics and Telecommunications Engineering Lnicst

Recently, there has been an increasing demand for traffic simulation and congestion prediction for urban planning, especially for infection simulation due to the Covid-19 epidemic. On the other hand, the widespread use of wearable devices has made it possible to collect a large amount of user location history with high accuracy, and it is expected that this data will be used for simulation. However, it is difficult to collect location histories for the entire population of a city, and detailed data that can reproduce trajectories is expensive. In addition, such personal location histories contain private information such as addresses and workplaces, which restricts the use of raw data. This paper proposes Agent2Vec, a mobility modeling model based on unsupervised learning. Using this method, we generate synthetic human flow data without personal information.

DOI： 10.1007/978-3-030-94822-1_36

その他リンク： https://dblp.uni-trier.de/db/conf/mobiquitous/mobiquitous2021.html#TamuraUAYK21

Smartphone Localization with Solar-Powered BLE Beacons in Warehouse 査読有り

Kano, K; Yoshida, T; Hayashida, N; Asai, Y; Matsuyama, H; Katayama, S; Urano, K; Yonezawa, T; Kawaguchi, N

DISTRIBUTED, AMBIENT AND PERVASIVE INTERACTIONS, SMART ENVIRONMENTS, ECOSYSTEMS, AND CITIES, PT I 13325 巻頁： 291 - 310 2022年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics

Workloads in logistics warehouses have been increasing to meet growing demand, and a labor shortage has become a problem. Utilizing information of laborer locations leads to an increase in productivity. We propose an integrated positioning method using solar-powered Bluetooth Low Energy (BLE) beacons. They are easy to install and maintenance-free since they can work without power sources. However, their advertisement interval depends on illuminance and is unstable. Moreover, there are many obstructions in warehouses, such as shelves and products, which cause signal attenuation, interference, and packet losses. We apply particle filters, map matching, and speed prediction with a neural network model to improve robustness and accuracy. We installed 94 beacons in a logistics warehouse. We evaluated the accuracy and found that our method is more accurate than a baseline method.

DOI： 10.1007/978-3-031-05463-1_21

その他リンク： https://dblp.uni-trier.de/db/conf/hci/hci2022-24.html#KanoYHAMKUYK22

大規模屋外施設におけるWi-Fiパケットセンサへの影響と利活用の検証査読有り Open Access

米澤拓郎

情報処理学会論文誌デジタルプラクティス 3 巻頁： 12 - 21 2022年

Deep Learning for Ballroom Dance Recognition: A Temporal and Trajectory-Aware Classification Model With Three-Dimensional Pose Estimation and Wearable Sensing 査読有り

Matsuyama, H; Aoki, S; Yonezawa, T; Hiroi, K; Kaji, K; Kawaguchi, N

IEEE SENSORS JOURNAL 21 巻 ( 22 ) 頁： 25437 - 25448 2021年11月

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：IEEE Sensors Journal

Dance performance recognition methods have been investigated and shown various applications such as picture-pose evaluation and synchronizing foot timing and direction. However, detailed analysis and feedback are still missing. To provide them, understanding the performance by component level is necessary. Specifically, we formulate it as a dance-figure classification problem using three-dimensional body joints and wearable sensors. Our model is based on long short-term memory (LSTM) and includes the temporal and trajectory-wise structure that uses the trajectory information in a timestep and the temporal masking module. As a result, we achieved 93% accuracy with our proposed method, which is highly overwhelming the baseline result (84.7%) and very close to the accuracy of the experienced dancers (93.6%). We have made the dataset of ballroom dance performance dataset open to researchers to develop the activity recognition field further.

DOI： 10.1109/JSEN.2021.3098744

An end-to-end ble indoor localization method using lstm

Urano K., Hiroi K., Yonezawa T., Kawaguchi N.

Journal of Information Processing 29 巻 ( 0 ) 頁： 58 - 69 2021年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：一般社団法人情報処理学会

This paper proposes an indoor localization method for Bluetooth Low Energy (BLE) devices using an end-to-end LSTM neural network. We focus on a large-scale indoor space where there is a tough environment for wireless indoor localization due to signal instability. Our proposed method adopts end-to-end localization, which means input is a time-series of signal strength and output is the estimated location at the latest time in the input. The neural network in our proposed method consists of fully-connected and LSTM layers. We use a custom-made loss function with 3 error components: MSE, the direction of travel, and the leap of the estimated location. Considering the difficulty of data collection in a short preparation term, the data generated by a simple signal simulation is used in the training phase, before training with a small amount of real data. As a result, the estimation accuracy achieves an average of 1.92 m, using the data collected in GEXPO exhibition in Miraikan, Tokyo. This paper also evaluates the estimation accuracy assuming the troubles in a real operation.

DOI： 10.2197/ipsjjip.29.58

A Basic Study on Ballroom Dance Figure Classification with LSTM Using Multi-modal Sensor 査読有り

Matsuyama, H; Hiroi, K; Kaji, K; Yonezawa, T; Kawaguchi, N

ACTIVITY AND BEHAVIOR COMPUTING, ABC 2020 204 巻頁： 209 - 226 2021年

記述言語：英語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：Smart Innovation Systems and Technologies

The paper presents a ballroom dance figure classification method with LSTM using video and wearable sensors. Ballroom dance is a popular sport among people regardless of age or sex. However, learning ballroom dance is very difficult for less experienced dancers as it has many complex types of “dance figures”, which is a completed set of footsteps. Therefore, we aim to develop a system to assist dance exercise which gives advice proper to each dance figure characteristic by recognizing dance figures correctly. While the major approach to recognize dance performance is to utilize video, we cannot simply adopt it for ballroom dance because the images of dancers overlap each other. To solve the problem, we propose a hybrid figure recognition method combining video and wearable sensors to enhance its accuracy and robustness. We collect video and wearable sensor data of seven dancers including acceleration, angular velocity, and body parts location change by pose estimation. After that, we preprocess them and put them into an LSTM-based deep learning network. As a result, we confirmed that our approach achieved an F1-score of 0.86 for 13 figure types recognition using the multi-modal sensors with trial-based fivefold cross-validation. We also performed user-based cross-validation, and sliding window algorithms. In addition, we compared the results with our previous method using Random Forest and also evaluated the robustness with occlusions. We found the LSTM-based method worked better than Random Forest with keypoint data. On the other hand, LSTM could not perform well with a sliding window algorithm. We consider the LSTM-based method would work better with a larger dance figure data, which is our next work. In addition, we will investigate how to solve occlusion problems with pose estimation.

DOI： 10.1007/978-981-15-8944-7_13

A Gaze-Based Unobstructive Information Selection by Context-Aware Moving UI in Mixed Reality 査読有り

Hayashida, N; Matsuyama, H; Aoki, S; Yonezawa, T; Kawaguchi, N

DISTRIBUTED, AMBIENT AND PERVASIVE INTERACTIONS, DAPI 2021 12782 巻頁： 301 - 315 2021年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics

Mixed reality (MR) technology has been attracting attention in the automobile industry and logistics industry for work training and remote work support for newcomers. However, work support using MR technology has the problem that displaying too much information obstructs the user’s field of vision and rather interferes with the work. Therefore, it is necessary to detect the user’s intention and provide only the information that the user wants. In addition, the system should be able to detect naturally from the user’s behavior without interrupting the work, which the user explicitly selects. To solve these problems, in this paper, we use the user’s gaze to determine whether to display content by estimating whether or not they are looking at information. Another problem with gaze-based research in MR work support is that it is difficult to determine which space you are looking at in a space where virtual space and real space intersect. In order to overcome this problem, in this research, we grasp the user’s behavior from the movement of the user’s gaze, and we suggest a way to determine make the user interface (UI) the movement of the gaze that is unlikely to occur during that behavior to see if the user is looking at the UI. As a result of the experiment, we were able to promote the movement of gaze that is different from the characteristics of the movement of gaze during work with the proposed UI movement.

DOI： 10.1007/978-3-030-77015-0_22

その他リンク： https://dblp.uni-trier.de/db/conf/hci/hci2021-21.html#HayashidaMAYK21

Attention-Guidance Method Based on Conforming Behavior of Multiple Virtual Agents for Pedestrians 査読有り

Yoshida, N; Yonezawa, T

PROCEEDINGS OF THE 21ST ACM INTERNATIONAL CONFERENCE ON INTELLIGENT VIRTUAL AGENTS (IVA) 頁： 209 - 215 2021年

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：Proceedings of the 21st ACM International Conference on Intelligent Virtual Agents Iva 2021

Humans observe the behavior of others and exhibit conforming behavior according to the situation. In this study, we propose a method to change the user's walking behavior and attention by changing the ratio of multiple virtual agents in action. We verified whether the attention to a poster on the wall and the walking behavior of a pedestrian changed according to the type and proportion of actions exhibited by multiple agents presented to a particular pedestrian. The results showed that the participant's walking speed and attention to the wall advertisement changed according to the agents' walking behavior, such as deceleration and stopping.

DOI： 10.1145/3472306.3478346

RGNet: Robust Gravity Estimation Neural Network for IMU-based Localization Using Smartphone 査読有り

Yoshida T., Urano K., Aoki S., Yonezawa T., Kawaguchi N.

13th International Conference on Mobile Computing and Ubiquitous Network Icmu 2021 頁： 1 - 8 2021年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：13th International Conference on Mobile Computing and Ubiquitous Network Icmu 2021

With the rapid development of Micro Electro-Mechanical Systems (MEMS) technologies, indoor navigation and localization with Inertial Measurement Unit (IMU) has been increasingly feasible. IMU-based indoor localization is a low-cost, energy-efficient, and infrastructure-free approach. There are various methods for it, and most of them require gravity estimation (e.g. the projection of angular velocity, the extraction of horizontal acceleration). In particular, when you use a smartphone, it changes the orientation of its IMU frequently, therefore the gravity estimation needs to be more robust to sensor orientation and its noise. In this paper, we propose a gravity estimation method based on deep learning called RGNet (Robust Gravity Estimation Neural Network) that is robust to sensor orientation and noise. We train an LSTM (Long short-term memory)-based neural network that estimates gravity from acceleration and angular velocity. Furthermore, for the problem that it is difficult to prepare the ground truth of gravity directly, we propose a method to train the gravity estimation neural network indirectly using the heading, taking advantage of the fact that the heading can be estimated from the gravity and angular velocity. The evaluation results show that the accuracy of the proposed method outperformed the baseline gravity estimation method (Android API, Low-pass filter, and Extended Kalman filter). We also confirmed that the accuracy of IMU-based localization is affected by the difference in gravity estimation methods.

DOI： 10.23919/ICMU50196.2021.9638853

その他リンク： https://dblp.uni-trier.de/db/conf/icmu/icmu2021.html#YoshidaUAYK21

RGNet: Robust Gravity Estimation Neural Network for IMU-based Localization Using Smartphone 査読有り

Yoshida, T; Urano, K; Aoki, S; Yonezawa, T; Kawaguchi, N

13TH INTERNATIONAL CONFERENCE ON MOBILE COMPUTING AND UBIQUITOUS NETWORK (ICMU2021) 2021年

Estimating and Leveraging Latent Social Demand Based on IoT sensors: An Empirical Study in a Large Public Park 査読有り

Nagata Y., Murai D., Katayama S., Urano K., Aoki S., Yonezawa T., Kawaguchi N.

13th International Conference on Mobile Computing and Ubiquitous Network Icmu 2021 頁： 1 - 8 2021年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：13th International Conference on Mobile Computing and Ubiquitous Network Icmu 2021

The continuously changing conditions of cities are now technically understandable in the information space through real-world sensing methods and analytical methods such as big data analysis and machine learning. On the other hand, it is currently difficult to estimate and present what the people in the city need (latent demand). This paper aims to solve latent demand by developing Latent Demand Resolver (LD-Resolver), the new latent demand estimation and exchanging system. LD-Resolver has two components, Latent Demand Extractor (LD-Extractor) and Latent Demand Exchanger (LD-Exchanger). LD-Extractor extracts a latent demand from various social conditions using IoT sensors, Web, and SNS. LD-Exchanger has a new structure to exchange a latent demand with an appropriate service supply. Finally, we developed the LD-Resolver and conducted a demonstration experiment at the Higashiyama Zoo and Botanical Garden to verify the method. As a result of the two-week experiment, the proposed method can be effectively used in actual facility operations.

DOI： 10.23919/ICMU50196.2021.9638789

その他リンク： https://dblp.uni-trier.de/db/conf/icmu/icmu2021.html#NagataMKUAYK21

Estimating and Leveraging Latent Social Demand Based on IoT sensors: An Empirical Study in a Large Public Park 査読有り

Nagata, Y; Murai, D; Katayama, S; Urano, K; Aoki, S; Yonezawa, T; Kawaguchi, N

13TH INTERNATIONAL CONFERENCE ON MOBILE COMPUTING AND UBIQUITOUS NETWORK (ICMU2021) 2021年

DigiMobot: Digital Twin for Human-Robot Collaboration in Indoor Environments 査読有り

Fukushima, Y; Asai, Y; Aoki, S; Yonezawa, T; Kawaguchi, N

2021 32ND IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV) 2021-July 巻頁： 55 - 62 2021年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：IEEE Intelligent Vehicles Symposium Proceedings

Human-robot collaboration and cooperation are critical for Autonomous Mobile Robots (AMRs) in order to use them in indoor environments, such as offices, hospitals, libraries, schools, factories, and warehouses. Since a long transition period might be required to fully automate such facilities, we have to deploy AMRs while improving safety in the mixed environments of human and mobile robots. In addition, human behaviors in such environments might be difficult to predict. In this paper, we present a Digital Twin for Autonomous Mobile Robots system named DigiMobot to support, manage, monitor, and validate AMRs in indoor environments. First, DigiMobot can simulate human behaviors and robot movements to verify and validate AMRs to improve safety in a virtual world. Secondly, DigiMobot can monitor and manage AMRs in the physical world by collecting sensor data from each robot in real-time. Since DigiMobot enables us to test the robot systems in the virtual world, we can deploy and implement AMRs in each facility without any modifications. To show the feasibility of DigiMobot, we develop a software framework and two different types of autonomous mobile robots. Finally, we conduct real-world experiments in a warehouse located in Saitama, Japan, in which more than 400, 000 items are stored for commercial purposes.

DOI： 10.1109/IV48863.2021.9575499

その他リンク： https://dblp.uni-trier.de/db/conf/ivs/ivs2021.html#FukushimaAAYK21

Science and Technologies for Smart Cities - 6th EAI International Conference, SmartCity360° Virtual Event, December 2-4, 2020, Proceedings 査読有り

SmartCity360° 372 巻 2021年

Time-resolved X-ray photoelectron diffraction using an angle-resolved time-of-flight electron analyzer 査読有り

Ang A.K.R., Fukatsu Y., Kimura K., Yamamoto Y., Yonezawa T., Nitta H., Fleurence A., Yamamoto S., Matsuda I., Yamada-Takamura Y., Hayashi K.

Japanese Journal of Applied Physics 59 巻 ( 10 ) 2020年10月

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：Japanese Journal of Applied Physics

X-ray photoelectron diffraction (XPD) provides atomic resolution, element sensitive local structure information about the surfaces and interfaces of materials. In the work reported in this paper, a two-dimensional angle resolved time-of-flight (2DARTOF) system is used to perform time-resolved XPD experiments on epitaxial silicene. The two Si 2p peak components from the different atomic sites in the silicene layer allow extraction of the individual XPD patterns. Time-resolved measurements captured small angle shifts in the forward focusing peak, indicating laser-induced changes in the silicene structure. At 10 ns, the XPD patterns appear to relax back to the equilibrium state. This work demonstrates that 2DARTOF systems are well suitable for time-resolved XPD measurements.

DOI： 10.35848/1347-4065/abb57e

CityFlow: Supporting Spatial-Temporal Edge Computing for Urban Machine Learning Applications 査読有り

Kawano, M; Yonezawa, T; Tanimura, T; Giang, NK; Broadbent, M; Lea, R; Nakazawa, J

3RD EAI INTERNATIONAL CONFERENCE ON IOT IN URBAN SPACE 頁： 3 - 15 2020年

記述言語：英語掲載種別：論文集(書籍)内論文出版者・発行元：Eai Springer Innovations in Communication and Computing

A growing trend in smart cities is the use of machine learning techniques to gather city data, formulate learning tasks and models, and use these to develop solutions to city problems. However, although these processes are sufficient for theoretical experiments, they often fail when they meet the reality of city data and processes, which by their very nature are highly distributed, heterogeneous, and exhibit high degrees of spatial and temporal variance. In order to address those problems, we have designed and implemented an integrated development environment called CityFlow that supports developing machine learning applications. With CityFlow, we can develop, deploy, and maintain machine learning applications easily by using an intuitive data flow model. To verify our approach, we conducted two case studies: deploying a road damage detection application to help monitor transport infrastructure and an automatic labeling application in support of a participatory sensing application. These applications show both the generic applicability of our approach, and its ease of use; both critical if we wish to deploy sophisticated ML based applications to smart cities.

DOI： 10.1007/978-3-030-28925-6_1

Poster Abstract: SelfGuard: Semi-Automated Activity Tracking for Enhancing Self-Protection against the COVID-19 Pandemic 査読有り

Nishiyama, Y; Yonezawa, T; Sezaki, K

PROCEEDINGS OF THE 2020 THE 18TH ACM CONFERENCE ON EMBEDDED NETWORKED SENSOR SYSTEMS, SENSYS 2020 頁： 780 - 781 2020年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：Sensys 2020 Proceedings of the 2020 18th ACM Conference on Embedded Networked Sensor Systems

Contagious diseases like COVID-19 spread periodically and threaten our lives. Self-protection, such as washing hands, wearing a mask, and staying home, are simple and practical solutions to safeguard against these diseases. Most governments and health departments recommend that people maintain self-protection. Although continuous self-protection effectively prevents the spread of infection, only the intent to self-protect is unsustainable in the long term. In this study, we design, develop, and deploy an application to track users' daily activities semi-automatically and enhance self-protection behavior using mobile sensing and gamified feedback techniques. Currently, more than 324 people have installed the app via AppStore, and 52 users have shared their activity data to our research group.

DOI： 10.1145/3384419.3430592

その他リンク： https://dblp.uni-trier.de/db/conf/sensys/sensys2020.html#NishiyamaYS20

GANonymizer: Image Anonymization Method Integrating Object Detection and Generative Adversarial Network 査読有り

Tanimura, T; Kawano, M; Yonezawa, T; Nakazawa, J

3RD EAI INTERNATIONAL CONFERENCE ON IOT IN URBAN SPACE 頁： 109 - 121 2020年

記述言語：英語掲載種別：論文集(書籍)内論文出版者・発行元：Eai Springer Innovations in Communication and Computing

Sharing and analyzing image data from ubiquitous urban cameras must enable us to understand and predict various contexts of the city. Meanwhile, since such image data always contains privacy data such as people and cars, we cannot easily share and analyze the data through the Internet for the viewpoint of privacy protection. As a result, most of urban image data are only kept/shared within the camera owners, or even discarded to reduce risks of privacy data leakage. To solve the privacy problem and accelerate sharing of urban image data, we propose GANonymizer that automatically detects and removes objects related to privacy from the urban images. GANonymizer combines two neural networks: (1) a network which detects objects related to privacy such as persons and cars in an input image using object detection network and (2) a network that removes the detected objects naturally as though they are not exist originally. Through our experiment of applying GANonymizer to urban video images, we confirmed that GANonymizer partially achieved natural removal of objects related to privacy.

DOI： 10.1007/978-3-030-28925-6_10

Civic Crowdsensing Through Location-Aware Virtual Monsters 査読有り

Yonezawa, T; Sakamura, M; Kawaguchi, N; Nakazawa, J

DISTRIBUTED, AMBIENT AND PERVASIVE INTERACTIONS, DAPI 2020 12203 巻頁： 464 - 476 2020年

記述言語：日本語掲載種別：研究論文（国際会議プロシーディングス）出版者・発行元：Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics

We present a new model for encouraging people to get involved with monitoring and taking part in the life of cities. Cities could be smarter if IoT and people could serve as engaged and pro-active data resources (i.e., crowd sensing). This study tackles two challenges: methods by which the privacy of people who act as sensors/actuators can be guaranteed and methods to create a unified programming model for crowd sensors alongside other IoT functions. To achieve these goals, we introduce a new concept called Lokemon (Location Monster). Each sensing space is characterized as a personified target. Lokemon asks users to imagine themselves to be monsters associated with target spots when achieving sensing tasks. Lokemon is also expressed as a PubSub node so that the data from Lokemon can be easily accessed in the same way as data from IoT is assessed. The article explains the concept of Lokemon and its programming model. We report our evaluation of the effectiveness of Lokemon in a campus experiment that was performed for four weeks.

DOI： 10.1007/978-3-030-50344-4_33

その他リンク： https://dblp.uni-trier.de/db/conf/hci/hci2020-23.html#YonezawaSKN20

　前のページ - 次のページ　

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16