| 1. |
Yugo Nakamura, Rei Nakaoka, Yuki Matsuda, Keiichi Yasumoto, eat2pic: An Eating-Painting Interactive System to Nudge Users into Making Healthier Diet Choices, Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 10.1145/3580784, 7, 1, 1-23, 2023.03, Given the complexity of human eating behaviors, developing interactions to change the way users eat or their choice of meals is challenging. In this study, we propose an interactive system called eat2pic designed to encourage healthy eating habits such as adopting a balanced diet and eating more slowly, by refraining the task of selecting meals into that of adding color to landscape pictures. The eat2pic system comprises a sensor-equipped chopstick (one of a pair) and two types of digital canvases. It provides fast feedback by recognizing a user's eating behavior in real time and displaying the result on a small canvas called "one-meal eat2pic."Moreover, it also provides slow feedback by displaying the number of colors of foods that the user consumed on a large canvas called "one-week eat2pic."The former was designed and implemented as a guide to help people eat more slowly, and the latter to encourage people to select more balanced menus. Through two user studies, we explored the experience of interaction with eat2pic, in which users' daily eating behavior was reflected in a series of "paintings,"that is, images produced by the automated system. The experimental results suggest that eat2pic may provide an opportunity for reflection in meal selection and while eating, as well as assist users in becoming more aware of how they are eating and how balanced their daily meals are. We expect this system to inspire users' curiosity about different diets and ways of eating. This research also contributes to expanding the design space for products and services related to dietary support.. |
| 2. |
Yugo Nakamura, Yutaka Arakawa, Takuya Kanehira, Masashi Fujiwara, Keiichi Yasumoto, SenStick: Comprehensive Sensing Platform with an Ultra Tiny All-In-One Sensor Board for IoT Research, Journal of Sensors, 10.1155/2017/6308302, 2017, 6308302-16, 2017.09, We propose a comprehensive sensing platform called SenStick, which is composed of hardware (ultra tiny all-in-one sensor board), software (iOS, Android, and PC), and 3D case data. The platform aims to allow all the researchers to start IoT research, such as activity recognition and context estimation, easily and efficiently. The most important contribution is the hardware that we have designed. Various sensors often used for research are embedded in an ultra tiny board with the size of 50 mm (W) × 10 mm (H) × 5 mm (D) and weight around 3 g including a battery. Concretely, the following sensors are embedded on this board: acceleration, gyro, magnetic, light, UV, temperature, humidity, and pressure. In addition, this board has BLE (Bluetooth low energy) connectivity and capability of a rechargeable battery. By using 110 mAh battery, it can run more than 15 hours. The most different point from other similar boards is that our board has a large flash memory for logging all the data without a smartphone. By using SenStick, all the users can collect various data easily and focus on IoT data analytics. In this paper, we introduce SenStick platform and some case studies. Through the user study, we confirmed the usefulness of our proposed platform.. |
| 3. |
Yohei Torigoe, Yugo Nakamura, Manato Fujimoto, Yutaka Arakawa, Keiichi Yasumoto, Strike activity detection and recognition using inertial measurement unit towards kendo skill improvement support system, Sensors and Materials, 10.18494/SAM.2020.2615, 32, 2, 651-673, 2020.01, In the field of sports, there are increasing opportunities to use inertial measurement units (IMUs) to enhance the training process and improve the performance of athletes. We focus on kendo, a traditional martial art using shinai (bamboo swords) in Japan, and propose methods for detecting and recognizing strike activities using IMUs towards realizing a kendo skill improvement support system. We used a sensor data set of strike activities obtained from 14 participants (seven kendo-experienced and seven inexperienced persons). We attached four IMUs to the participants’ right wrist, waist, and shinai (tsuba and saki-gawa). First, to detect the strike activity, we calculated the dynamic time warping (DTW) distance between the training data and the time series data, and detected the strike activity sections. The proposed method detected strike activities with a high accuracy of 95.0%. Next, to recognize the strike activity, we recognized five types (Center-Men, Right-Men, Left-Men, Dō, and Kote). In the person-dependent (PD) case, we achieved an accuracy of 89.5% using data of the right wrist. In the person-independent (PI) case, we achieved an accuracy of 54.9% using IMUs attached to the three positions. These results clarified the points to be improved in the proposed method to realize the support system.. |
| 4. |
Ko Watanabe, Yusuke Soneda, Yuki Matsuda, Yugo Nakamura, Yutaka Arakawa, Andreas Dengel, Shoya Ishimaru, DisCaaS: Micro Behavior Analysis on Discussion by Camera as a Sensor, Sensors, 10.3390/s21175719, 21, 17, 5719-5719, 2021.08, The emergence of various types of commercial cameras (compact, high resolution, high angle of view, high speed, and high dynamic range, etc.) has contributed significantly to the understanding of human activities. By taking advantage of the characteristic of a high angle of view, this paper demonstrates a system that recognizes micro-behaviors and a small group discussion with a single 360 degree camera towards quantified meeting analysis. We propose a method that recognizes speaking and nodding, which have often been overlooked in existing research, from a video stream of face images and a random forest classifier. The proposed approach was evaluated on our three datasets. In order to create the first and the second datasets, we asked participants to meet physically: 16 sets of five minutes data from 21 unique participants and seven sets of 10 min meeting data from 12 unique participants. The experimental results showed that our approach could detect speaking and nodding with a macro average f1-score of 67.9% in a 10-fold random split cross-validation and a macro average f1-score of 62.5% in a leave-one-participant-out cross-validation. By considering the increased demand for an online meeting due to the COVID-19 pandemic, we also record faces on a screen that are captured by web cameras as the third dataset and discussed the potential and challenges of applying our ideas to virtual video conferences.. |
| 5. |
Yugo Nakamura, Yutaka Arakawa, Keiichi Yasumoto, Smart experimental platform for collecting various sensing data from various things, UbiComp 2016 Adjunct - Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing, 10.1145/2968219.2968268, 1751-1754, 2016.09, In this paper, we report our efforts to recognize human context through the things sensorized by our sensor platform As you know, many things wearing on your body have alread become a sensor to understand our activities. Mos famous things are watches and glasses. In the future, w suppose that unwearable things, such as cutlery, sport gea and instruments, would be also a sensor for recognizing ou activities. Compared with our movements, the movement o unwearable things will have more variation. Also, suitabl sensor and proper algorithm would be different according t the things. To enhance the senserization of various things we have proposed a platform, called SenStick, that consist of ultra tiny all-in-one sensor board and sophisticate mobile app. If using it, every researcher can try embeddin various sensor into various things and can focus on dat analysis. We explain the possibility of smart experimen platform with introducing various sensorized things on th platform.. |
| 6. |
Yugo Nakamura, Yutaka Arakawa, Keiichi Yasumoto, Smart Experiment Platform for Wearable Computing, First Workshop on Eye Wear Computing collocated with ISWC/UbiComp (EYEWEAR 2016), 2016.09. |
| 7. |
Yugo Nakamura, Yutaka Arakawa, Keiichi Yasumoto, Senstick: A rapid prototyping platform for sensorizing things, 2016 9th International Conference on Mobile Computing and Ubiquitous Networking, ICMU 2016, 10.1109/ICMU.2016.7742087, 31-36, 2016.11, In this paper, we propose a novel rapid prototyping platform called SenStick, which is composed of both hardware and software. The main purpose of our platform is to sensorize our personal belongings easily and smart for recognizing our living activities. The most impressive point is its size. The size is 75mm(W) x 10mm(H) x 5mm(D) and its weight is around 3 (g) including a battery. On this tiny board, 8 sensors (acceleration, gyro, magnetic, light, UV, temperature, humidity, and pressure), flash memory, BLE, and battery are embedded in high density. The battery life in stand-alone mode is more than 12 hours. Second interesting point is the support software for iOS and Android OS. It can monitor the sensing data as well as can record the ground truth video simultaneously and synchronously. Furthermore, 3D CAD data of various case designs will be open for users in SenStick community site. As a result, SenStick enables everyone to sense every activities easily and smart.. |
| 8. |
Yugo Nakamura, Takuya Kanehira, Yutaka Arakawa, Keiichi Yasumoto, SenStick 2: Ultra tiny all-in-one sensor with wireless charging, UbiComp 2016 Adjunct - Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing, 10.1145/2968219.2971399, 337-340, 2016.09, In this demo, we present a novel rapid prototyping platform called SenStick2, which is all-in-one sensor with wireless charging. The main purpose of our platform is to sensorize our personal belongings easily and smartly for recognizing our living activities. The size of SenStick2 hardware (board) is 50mm(W) 10mm(H) 5mm(D) and its weight is around 3 (g) including a battery. On this tiny board, 8 sensors (acceleration, gyro, magnetic, light, UV, temperature, humidity, and pressure), flash memory, BLE, and battery are embedded in high density. The battery life in stand-Alone mode is more than 15 hours. For SenStick platform, we have developed the support software for iOS and Android OS. It can monitor the sensing data and also record the ground truth video simultaneously and synchronously. Furthermore, 3D CAD data of various case designs will be open for users in SenStick community site. By providing 3D case data, all the researcher can measure every tiny activities in a daily life easily and smartly.. |
| 9. |
Yugo Nakamura, Hirohiko Suwa, Yutaka Arakawa, Hirozumi Yamaguchi, Keiichi Yasumoto, Middleware for Proximity Distributed Real-Time Processing of IoT Data Flows, Proceedings - International Conference on Distributed Computing Systems, 10.1109/ICDCS.2016.101, 2016-August, 771-772, 2016.08, EdgeComputing and Fog Computing are new paradigms where data processing is executed in or on the edge of networks to mitigate cloud server load. However, EdgeComputing and Fog Computing still need powerful servers on the edge of networks which impose additional costs for deployments. We proposed a platform called IFoT (Information Flow of Things) that efficiently performs distributed processing as well as distribution and analysis of data streams near their sources based on "Process On Our Own (PO3)" concept. In IFoT, processing of tasks for cloud servers is delegated to an ad-hoc distributed system consisting of proximity IoT devices for distributed real-time stream processing. In this demonstration, we show a face recognition system for person tracking developed on top of IFoT middleware which locally processes video streams in real-time and in a distributed manner by using computational resources of IoT devices.. |
| 10. |
Yugo Nakamura, Hirohiko Suwa, Yutaka Arakawa, Hirozumi Yamaguchi, Keiichi Yasumoto, Design and Implementation of Middleware for IoT Devices toward Real-Time Flow Processing, Proceedings - 2016 IEEE 36th International Conference on Distributed Computing Systems Workshops, ICDCSW 2016, 10.1109/ICDCSW.2016.37, 162-167, 2016.11, Thanks to rapid advance and penetration of IoT devices, it is becoming possible to sense almost every information of real-world. This urges us to utilize data streams continuously generated from IoT devices in real-Time. In this paper, aiming to locally process data streams by using computational resources of IoT devices, we propose middleware for IoT devices where the devices process data streams in real-Time and in a distributed manner. The proposed middleware provides four functions: (1) distribution of tasks issued by application software into sub-Tasks and distributed execution of the sub-Tasks over multiple IoT devices, (2) distribution of data streams over IoT devices, (3) real-Time analysis of the data streams, and (4) seamless integration of sensors and actuators. We have implemented a prototype of the proposed middleware for Raspberry Pi and show its basic performance.. |
| 11. |
Masato Hidaka, Yuki Matsuda, Shogo Kawanaka, Yugo Nakamura, Manato Fujimoto, Yutaka Arakawa, Keiichi Yasumoto, A System for Collecting and Curating Sightseeing Information toward Satisfactory Tour Plan Creation, The Second International Workshop on Smart Sensing Systems (IWSSS’17), 2017.08. |
| 12. |
Kazuhito Umeki, Yugo Nakamura, Manato Fujimoto, Yutaka Arakawa, Keiichi Yasumoto, Real-Time Congestion Estimation in Sightseeing Spots with BLE Devices, 2018 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2018, 10.1109/PERCOMW.2018.8480395, 430-432, 2018.10, Recently, there is a growing demand to know con- gestion information on sightseeing spots in real-time to provide a satisfactory tour plan to tourists. Many studies on a congestion estimation have been conducted so far. However, most of them suffer from high deployment/operation costs and/or rely on contributions by users with smartphones/sensors. In this paper, we propose a novel system that estimates congestion of sightseeing spots in real-time without attaching any device to tourists by observing the distribution of per-RSSI intensity occurrences in each time window when beacon signals are periodically sent between BLE (Bluetooth Low Energy) transceivers installed in sightseeing spots. In other words, our system can estimate the congestion degree in sightseeing spots simply by using the property of RSSI intensity which dramatically changes depending on the number of people. Therefore, the proposed system is simple and low cost and can estimate the congestion easily without any special devices attached to tourists. In the demonstration, we show the congestion degree in three levels (low, medium, and high) changing in real-time depending on the number of audience in the demonstration site.. |
| 13. |
Yutaka Arakawa, Yugo Nakamura, Hirohiko Suwa, Yoshinori Umetsu, Manato Fujimoto, Keiichi Yasumoto, Poster: Feasibility study toward a battery-free place recognition system based on solar cells, UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers, 10.1145/3267305.3267624, 1-4, 2018.10, In this paper, we propose a battery-free place recognition system that utilizes solar-cells as a sensor for localization. Our system combines multiple solar cells having different characteristics against the light environment. As an initial work, we select five kinds of solar cells available in the market and investigate the characteristics of them. Then, we show a potential for estimating the place based on the variation of electricity amounts generated from solar cells. Finally, we show that our proposed system distinguish nine places with 88.0% accuracy.. |
| 14. |
Yugo Nakamura, Yoshinori Umetsu, Jose Paolo Talusan, Keiichi Yasumoto, Wataru Sasaki, Masashi Takata, Yutaka Arakawa, Multi-stage activity inference for locomotion and transportation analytics of mobile users, UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers, 10.1145/3267305.3267526, 1579-1588, 2018.10, In this paper, we, Ubi-NUTS Japan, introduce a multi-stage activity inference method that can recognize a user's mode of locomotion and transportation using mobile device sensors. We use the Sussex-Huawei Locomotion-Transportation (SHL) dataset to tackle the SHL recognition challenge, where the goal is to recognize 8 modes of locomotion and transportation (still, walk, run, bike, car, bus, train, and subway) activities from the inertial sensor data of a smartphone. We adopt a multi-stage approach where the 8 class classification problem is divided into multiple sub-problems considering the similarity of each activity. Multimodal sensor data collected from a mobile phone are inferred using a proposed pipeline that combines feature extraction and 4 different types of classifiers generated using the random forest algorithm. We evaluated our method using data from over 271 hours of daily activities of 1 participant and the 5-fold cross-validation. Evaluation results demonstrate that our method clearly recognizes the 8 types of activities with an average F1-score of 97%.. |
| 15. |
Masashi Takata, Manato Fujimoto, Keiichi Yasumoto, Yugo Nakamura, Yutaka Arakawa, Investigating the capitalize effect of sensor position for training type recognition in a body weight training support system, UbiComp/ISWC 2018 - Adjunct Proceedings of the 2018 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2018 ACM International Symposium on Wearable Computers, 10.1145/3267305.3267504, 1404-1408, 2018.10, A body weight training (BWT) means the training which utilizes the self-weight instead of the weight machine. The feedback of form and proper training menu recommendation is important for maximizing the effect of BWT. The objective of this study is to realize a novel support system which allows beginners to perform effective BWT alone, under wearable computing environment. To make an effective feedback, it is necessary to recognize BWT type with high accuracy. However, since the accuracy is greatly affected by the position of wearable sensors, we need to know the sensor position which achieves the high accuracy in recognizing the BWT type. We investigated 10 types BWT recognition accuracy for each sensor position. We found that waist is the best position when only 1 sensor is used. When 2 sensors are used, we found that the best combination is of waist and wrist. We conducted an evaluation experiment to show the effectiveness of sensor position. As a result of leave-one-person-out cross-validation from 13 subjects to confirm validity, we calculated the F-measure of 93.5% when sensors are placed on both wrist and waist.. |
| 16. |
Yugo Nakamura, Teruhiro Mizumoto, Hirohiko Suwa, Yutaka Arakawa, Hirozumi Yamaguchi, Keiichi Yasumoto, In-situ resource provisioning with adaptive scale-out for regional IoT services, Proceedings - 2018 3rd ACM/IEEE Symposium on Edge Computing, SEC 2018, 10.1109/SEC.2018.00022, 203-213, 2018.12, In an era where billions of IoT devices have been deployed, edge/fog computing paradigms are attracting attention for their ability to reduce processing delays and communication overhead. In order to improve Quality of Experience (QoE) of regional IoT services that create timely geo-spatial information in response to users’ queries, it is important to efficiently allocate sufficient resources based on the computational demand of each service. However since edge/fog devices are assumed to be heterogeneous (in terms of their computational power, network performance to other devices, deployment density, etc.), provisioning computational resources according to computational demand becomes a challenging constrained optimization problem. In this paper, we formulate a delay constrained regional IoT service provisioning (dcRISP) problem. dcRISP assigns computational resources of devices based on the demand of the regional IoT services in order to maximize users’ QoE. We also present dcRISP+, an extension of dcRISP, that enables resource selection to extend beyond the initial area in order to satisfy increasing computational demands. We propose a provisioning algorithm, in-situ resource area selection with adaptive scale out and in-situ task scheduling based on a tabu search, to solve the dcRISP+ problem. We conducted a simulation study of a tourist area in Kyoto where 4,000 IoT devices and 3 types of IoT services were deployed. Results show that our proposed algorithms can obtain higher user QoE compared to conventional resource provisioning algorithms.. |
| 17. |
Evacuation Center Determination Method Considering Bias of Congestion Degree on Areas. |
| 18. |
Yugo Nakamura, Teruhiro Mizumoto, Hirohiko Suwa, Yutaka Arakawa, Hirozumi Yamaguchi, Keiichi Yasumoto, Design and Evaluation of In-Situ Resource Provisioning Method for Regional IoT Services, 2018 IEEE/ACM 26th International Symposium on Quality of Service, IWQoS 2018, 10.1109/IWQoS.2018.8624127, 1-2, 2019.01, In an era where billions of IoT devices are deployed, edge/fog computing paradigms are attracting attention for their ability to reduce processing delays and mitigate waste of communication resources. However, since the computing system assumed by edge/fog paradigms have heterogeneity (in terms of the computing power of devices, network performance between devices, device density, etc.), provisioning computational resources according to computational demand becomes a challenging constrained optimization problem. In this paper, we propose in-situ resource provisioning method consisting of insitu resource area selection with adaptive scale out and in-situ task scheduling based on tabu search algorithm. We conducted a simulation study in a target regional area where 2,000 IoT devices and 10 IoT services are deployed to evaluate the effectiveness of the proposed algorithm. The simulation results show that our proposed algorithm can obtain higher user QoS compared to conventional resource provisioning algorithms.. |
| 19. |
Yoshinori Umetsu, Yugo Nakamura, Yutaka Arakawa, Manato Fujimoto, Hirohiko Suwa, EHAAS: Energy harvesters as a sensor for place recognition on wearables, 2019 IEEE International Conference on Pervasive Computing and Communications, PerCom 2019, 10.1109/PERCOM.2019.8767385, abs/1903.08592, 1-10, 2019.03, A wearable based long-term lifelogging system is desirable for the purpose of reviewing and improving users lifestyle habits. Energy harvesting (EH) is a promising means for realizing sustainable lifelogging. However, present EH technologies suffer from instability of the generated electricity caused by changes of environment, e.g., the output of a solar cell varies based on its material, light intensity, and light wavelength. In this paper, we leverage this instability of EH technologies for other purposes, in addition to its use as an energy source. Specifically, we propose to determine the variation of generated electricity as a sensor for recognizing "places" where the user visits, which is important information in the lifelogging system. First, we investigate the amount of generated electricity of selected energy harvesting elements in various environments. Second, we design a system called EHAAS (Energy Harvesters As A Sensor) where energy harvesting elements are used as a sensor. With EHAAS, we propose a place recognition method based on machine-learning and implement a prototype wearable system. Our prototype evaluation confirms that EHAAS achieves a place recognition accuracy of 88.5% F-value for nine different indoor and outdoor places. This result is better than the results of existing sensors (3-axis accelerometer and brightness). We also clarify that only two types of solar cells are required for recognizing a place with 86.2% accuracy.. |
| 20. |
Jose Paolo Talusan, Francis Tiausas, Sopicha Stirapongsasuti, Yugo Nakamura, Teruhiro Mizumoto, Keiichi Yasumoto, Evaluating Performance of In-Situ Distributed Processing on IoT Devices by Developing a Workspace Context Recognition Service, 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019, 10.1109/PERCOMW.2019.8730693, 633-638, 2019.03, With the number of IoT devices expected to exceed 50 billion in 2023, edge and fog computing paradigms are beginning to attract attention as a way to process the massive amounts of raw data being generated. However, these paradigms do not consider the processing capabilities of the existing commodity IoT devices in the wild. In order to solve this challenge, we are developing a new middleware platform called IFoT, which processes various sensor data while considering Quality of Service (QoS) by utilizing the computational resources of heterogeneous IoT devices within an area. This allows smart services to be created and processed in parallel by various IoT devices. In this paper, we show the effectiveness of the IFoT based approach of constructing services. We designed and implemented a workspace context recognition service, utilizing environmental sensor data processed in a distributed manner according to the IFoT framework. We evaluate the QoS of IFoT middleware and its feasibility when used on commodity devices such as the Raspberry Pi, through the service.. |
| 21. |
Masashi Takata, Yugo Nakamura, Yohei Torigoe, Manato Fujimoto, Yutaka Arakawa, Keiichi Yasumoto, Strikes-Thrusts Activity Recognition Using Wrist Sensor Towards Pervasive Kendo Support System, 2019 IEEE International Conference on Pervasive Computing and Communications Workshops, PerCom Workshops 2019, 10.1109/PERCOMW.2019.8730861, 243-248, 2019.03, In this paper, we focus on Kendo, which is a traditional sport in Japan, and propose a strikes-thrusts activity recognition method using a wrist sensor towards a pervasive Kendo support system. We collected the inertial sensor data set from 6 subjects. We attached 3 inertial sensor units (IMUs) on the subjects body, and 2 IMUs on the Shinai (bamboo sword used for Kendo). On the body, IMUs were placed on the Right Wrist, Waist and Right Ankle. On the Shinai, they were placed on the Tsuba and Saki-Gawa. We first classified strikes-thrusts activities consisting of 4 general types, Men, Tsuki, Do, and Kote, followed by further classification into 8 detailed types. We achieved 90.0% of F-measure in the case of 4-type classification and 82.6% of F-measure in the case of 8-type classification when learning and testing the same subjects data for only Right Wrist. Further, when adding data of sensors attached to the Waist and Right Ankle, we achieved 97.5% of F-measure for 4-type classification and 91.4% of F-measure for 8-type classification. As a result of leave-one-person-out cross-validation from 6 subjects to confirm generalized performance, in the case of 4-type classification, we achieved 77.5% of F-measure by using only 2 IMUs (Right Wrist and Shinai Tsuba).. |
| 22. |
Yugo Nakamura, Yuki Matsuda, Yutaka Arakawa, Keiichi Yasumoto, Waistonbelt x: A belt-type wearable device with sensing and intervention toward health behavior change, Sensors (Switzerland), 10.3390/s19204600, 19, 20, 4600-4600, 2019.10, Changing behavior related to improper lifestyle habits has attracted attention as a solution to prevent lifestyle diseases, such as diabetes, heart disease, arteriosclerosis, and stroke. To drive health behavior changes, wearable devices are needed, and they must not only provide accurate sensing and visualization functions but also effective intervention functions. In this paper, we propose a health support system, WaistonBelt X, that consists of a belt-type wearable device with sensing and intervention functions and a smartphone application. WaistonBelt X can automatically measure a waistline with a magnetometer that detects the movements of a blade installed in the buckle, and monitor the basic activities of daily living with inertial sensors. Furthermore, WaistonBelt X intervenes with the user to correct lifestyle habits by using a built-in vibrator. Through evaluation experiments, we confirmed that our proposed device achieves measurement of the circumference on the belt position (mean absolute error of 0.93 cm) and basic activity recognition (F1 score of 0.95) with high accuracy. In addition, we confirmed that the intervention via belt vibration effectively improves the sitting posture of the user.. |
