研究者詳細 - 河口　信夫

論文 - 河口　信夫

分割表示 563 件中 1 - 20 件目／全件表示 >>

Fixture-aware Panoramic Ceiling Image Synthesis Using an Omnidirectional Camera 招待有り査読有り Open Access

Daiki,Kohama, Yoshiteru,Nagata, Kazushige,Yasutake, Shin,Katayama, Kenta,Urano, Takuro,Yonezawa, Nobuo,Kawaguchi

Journal of Information Processing Vol.33 巻 ( 2025 ) 2025年3月

Life Pattern-Based Human Attribute Estimation Using Only GPS Data 査読有り国際誌 Open Access

IEEE ACCESS 13 巻頁： 131803 - 131822 2025年7月

担当区分：最終著者記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：IEEE Access

Analysis of large-scale GPS data enables the summarization and structuring of human mobility characteristics, providing insights into residents' behavior patterns. However, existing methods analyze sequences of individual-derived stay labels, such as home or workplace, which do not accurately reflect actual area functions and can fail to distinguish individuals based on differences in workplaces. This limitation arises because annotating stay locations and purposes in GPS data is challenging due to the complexity of urban structures and constraints on data quality. In this study, we propose ArLPH, a framework that takes only GPS data as input, models mobility from the perspective of 'when and what types of places he/she visited,' and enables the interpretation of life patterns. ArLPH consists of two modules: 'Area Modeling,' which represents urban areas based on usage patterns, and 'Human Modeling,' which represents individuals based on behavior patterns. By the processing flow of area modeling to human modeling, it becomes possible to analyze human mobility while considering area characteristics. Through an evaluation of a large-scale, real-world GPS dataset collected from tens of thousands of smartphone users, we demonstrate that ArLPH enables the characterization of individuals based on their life patterns and estimates attributes such as occupation (e.g., office workers or homemakers) and lifestyle (e.g., those who work frequently on weekends).

DOI： 10.1109/ACCESS.2025.3591811

Open Access

Web of Science

Scopus

その他リンク： https://ieeexplore.ieee.org/document/11091313

SemantiPack: An Efficient Real-World Data Compressor Using Structural and Semantic Metadata 査読有り

Nagata, Y; Kohama, D; Watanabe, Y; Katayama, S; Urano, K; Yonezawa, T; Kawaguchi, N

IEEE ACCESS 13 巻頁： 114159 - 114178 2025年6月

担当区分：最終著者記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：IEEE Access

The exponential growth of Real-World Data (RWD), primarily collected from IoT sensors and spanning domains such as mobility, environment, and energy consumption, presents critical challenges due to its scale, heterogeneity, and structural variability. Traditional compression methods often fail to adapt efficiently to these complexities, leading to sub-optimal storage and analysis performance. Additionally, while several metadata schemas exist to enhance the availability of RWD, smaller organizations often lack the resources to create and manage metadata effectively. This paper introduces RWD Profile, an automatically generatable metadata schema for RWD, and SemantiPack, which applies tailored compression to data fields in the individual RWD based on RWD Profile. RWD Profile has two kinds of metadata, Structural Profile and Semantic Profile. Structural Profile is generated through rule-based systems, while Semantic Profile is generated using large language models (LLMs) to capture semantic data properties. On the other hand, SemantiPack performs compression at the data field level in RWD using RWD Profile to achieve higher compression ratios. Experimental results demonstrate up to 23.2% improved compression rate for JSON and 19.3% for CSV compared to conventional methods while maintaining a faster or equivalent processing speed compared to conventional methods. Furthermore, SemantiPack supports lossy compression for applications prioritizing storage over precision. This research not only improves compression efficiency but also establishes a scalable solution for automated analysis and sustainable data utilization, paving the way for advancements in RWD management.

DOI： 10.1109/ACCESS.2025.3583829

Web of Science

Scopus

Digitization Methods for a Logistics Warehouse Towards Digital Twin-Driven Optimization

Kawaguchi N., Asai Y., Kano K., Takaki K., Mori Y., Suzuki Y., Watanabe K., Gushi Y., Katayama S., Urano K., Yonezawa T., Hashiguchi S.

Proceedings 2025 IEEE International Conference on Smart Computing Smartcomp 2025 頁： 186 - 193 2025年6月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Proceedings 2025 IEEE International Conference on Smart Computing Smartcomp 2025

The rapid growth of e-commerce, rising consumer expectations, and labor shortages in Japan pose challenges for logistics warehouse optimization. While automation has advanced outbound operations, inbound processes remain inefficient. This study presents a comprehensive approach to digitizing and optimizing large-scale logistics warehouses, with a case study at TRUSCO NAKAYAMA Corp. in partnership with Nagoya University. Key technologies include a large-scale camera array, multi-camera object tracking, and smartphone-based task estimation. Our contributions include real-time tracking of personnel and packages, cooperative annotation for improved object recognition, and synthetic data augmentation. Additionally, truck berth analysis and indoor localization enhance operational efficiency. To optimize worker shifts and warehouse layout, we apply Factorization Machine Quantum Annealing (FMQA), achieving a 37.4% reduction in lead times and a 14.3% decrease in labor hours. A visualization tool enables warehouse operators to make data-driven decisions. This research demonstrates the potential of digital transformation in logistics and provides a scalable framework for broader industry adoption.

DOI： 10.1109/SMARTCOMP65954.2025.00035

Scopus

その他リンク： https://db.uclab.jp/show/1538

Semi-Automated Framework for Digitalizing Multi-Product Warehouses with Large Scale Camera Arrays

Higashiura, K; Yokoyama, K; Asai, Y; Shimosato, H; Kano, K; Katayama, S; Urano, K; Yonezawa, T; Kawaguchi, N

2024 IEEE INTERNATIONAL CONFERENCE ON PERVASIVE COMPUTING AND COMMUNICATIONS, PERCOM 頁： 98 - 105 2024年3月

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：2024 IEEE International Conference on Pervasive Computing and Communications, PerCom 2024

As global demand for logistics continues to grow, optimizing the automation and efficiency of distribution warehouse operations is of paramount importance. Digitalizing warehouse environments, which refers to the process of sensing the physical space and extracting meaningful information from the obtained data, offers a promising solution to this challenge. However, converting raw warehouse data, such as video footage captured inside the warehouse, into actionable metadata (e.g., tracking the movement paths of workers and products or analyzing the usage patterns of different warehouse locations) often necessitates significant human intervention. The rise of machine learning further complicates this, as it requires the manual preparation of extensive training datasets. In this paper, we introduce a framework that semi-automates the digitalization process in complex warehouse settings. This framework employs dense optical flow and representation learning to autonomously segment warehouse objects and cluster similar objects, thereby substantially cutting down on annotation costs. To evaluate our approach, we constructed a large-scale data collection platform with over 60 fixed cameras in a real-world logistics warehouse, and the video data from this platform was then applied to our framework. Our evaluations indicate that our method markedly reduces both the time and resources required for warehouse digitalization using the captured video data.

DOI： 10.1109/PERCOM59722.2024.10494498

Web of Science

Scopus

都市を対象とした大規模移動履歴に基づく疑似人流データ生成手法招待有り査読有り Open Access

田村直樹, 浦野健太, 青木俊介, 米澤拓郎, 河口信夫

情報処理学会論文誌 64 巻 ( 1 ) 頁： 123 - 133 2023年1月

ER-Chat: A Text-to-Text Open-Domain Dialogue Framework for Emotion Regulation 査読有り

Katayama Shin, Aoki Shunsuke, Yonezawa Takuro, Okoshi Tadashi, Nakazawa Jin, Kawaguchi Nobuo

IEEE TRANSACTIONS ON AFFECTIVE COMPUTING 13 巻 ( 4 ) 頁： 2229 - 2237 2022年10月

担当区分：最終著者記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：IEEE Transactions on Affective Computing

Emotions are essential for constructing social relationships between humans and interactive systems. Although emotional and empathetic dialogue generation methods have been proposed for dialogue systems, appropriate dialogue involves not only mirroring emotions and always being empathetic but also complex factors such as context. This paper proposes Emotion Regulation Chat (ER-Chat) as an end-to-end dialogue framework for emotion regulation. Emotion regulation is concerned with actions to approach appropriate emotional states. Learning appropriate emotion and intent when responding on the basis of the context of the dialogue enables the generation of more human-like dialogue. We conducted automatic and human evaluations to demonstrate the superiority of ER-Chat over the baseline system. The results show that inclusion of emotion and intent prediction mechanisms enable generation of dialogues with greater fluency, diversity, emotion awareness, and emotion appropriateness, which are greatly preferred by humans.

DOI： 10.1109/TAFFC.2022.3191973

Web of Science

Scopus

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 Hitoshi, Aoki Shunsuke, Yonezawa Takuro, Hiroi Kei, Kaji Katsuhiko, Kawaguchi Nobuo

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

担当区分：最終著者記述言語：英語掲載種別：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

Web of Science

Scopus

大規模ユーザの滞在情報に基づくエリアの特徴付けとCOVID-19による影響分析査読有り Open Access

庄子和之, 青木俊介, 米澤拓郎, 河口信夫

情報処理学会論文誌 62 巻頁： 1644 - 1657 2021年10月

Real-time Activity-Aware Video Streaming While Preserving Privacy 査読有り

Ubicomp Companion 2025 Companion of the 2025 ACM International Joint Conference on Pervasive and Ubiquitous Computing 頁： 236 - 240 2025年12月

担当区分：最終著者記述言語：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：Ubicomp Companion 2025 Companion of the 2025 ACM International Joint Conference on Pervasive and Ubiquitous Computing

We present ActStream, a remote communication system that selectively shares the user and objects relevant to their current activities. By combining vision-language models and real-time segmentation, ActStream detects and highlights only activity-relevant elements, enhancing awareness while preserving privacy. The system identifies key objects through user focus or interaction, synthesizes them with the user's image, and streams the result to a remote partner. ActStream aims to provide useful context without exposing unnecessary information, highlighting the potential of selective sharing in privacy-sensitive remote collaboration.

DOI： 10.1145/3714394.3754399

Scopus

Smartphone-Based Activity Recognition in a Logistics Warehouse Using Self-supervised Representation Learning 査読有り

Ubicomp Companion 2025 Companion of the 2025 ACM International Joint Conference on Pervasive and Ubiquitous Computing 頁： 1067 - 1073 2025年12月

担当区分：最終著者記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Ubicomp Companion 2025 Companion of the 2025 ACM International Joint Conference on Pervasive and Ubiquitous Computing

To address labor shortages in logistics, digital twins that digitize physical environments are gaining attention, as they enable visualization and simulation-based optimization of warehouse operations. However, digitizing worker activities is particularly challenging due to operational burden and difficulty in adapting to diverse tasks. In response to this issue, we propose a deep learning-based framework for automatic warehouse task recognition using smartphone inertial sensors (accelerometer and gyroscope). Our model combines a CNN pretrained on unlabeled data using SimCLR, a self-supervised representation learning method, to improve generalization, and a Transformer encoder to capture task-level context. We collected data in a logistics warehouse and conducted cross-validation using 6 hours of annotated data from 10 individuals. As a result, the proposed method achieved a weighted F1 score of 0.89, improving by 7 points over the model without pretraining.

DOI： 10.1145/3714394.3756198

Scopus

Multi-City Next Location Prediction through Mobility-Derived Multi-Pattern Transfer Learning 査読有り

Terashima H., Tamura N., Shoji K., Urano K., Yonezawa T., Kawaguchi N.

33rd ACM Sigspatial International Conference on Advances in Geographic Information Systems ACM Sigspatial 2025 頁： 1218 - 1221 2025年11月

担当区分：最終著者記述言語：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：33rd ACM Sigspatial International Conference on Advances in Geographic Information Systems ACM Sigspatial 2025

In this study, we propose MoDeMIT (Mobility-Derived Multi-Insight Transfer), a human mobility prediction method that transfers various patterns shared across cities, extracted from mobility histories. Existing studies on human mobility prediction have primarily focused on learning the mobility patterns of users in the target city, without considering applications in cities with limited data. Moreover, relying solely on mobility patterns poses inherent limitations on prediction accuracy. The proposed MoDeMIT addresses these issues by defining and transferring multiple patterns shared across cities, such as lifestyle patterns and large-scale mobility patterns derived from mobility histories, as well as mobility patterns. This approach enables improvements in prediction accuracy compared to existing methods. We validate the effectiveness of MoDeMIT using real-world human mobility datasets. Furthermore, in the HuMob Challenge 2025 (GISCUP), MoDeMIT achieved a GEOBLEU score of [Average: 0.1632, CityA: 0.1504, CityB: 0.1471, CityC: 0.1801, CityD: 0.1753] and ranked within the top five teams.

DOI： 10.1145/3748636.3771311

Scopus

researchmap

Soccer Player Tracking Using Multiple Features for ID Switch Correction in Fixed-View Videos 査読有り

8TH INTERNATIONAL ACM WORKSHOP ON MULTIMEDIA CONTENT ANALYSIS IN SPORTS, MMSPORTS 2025 頁： 189 - 195 2025年10月

CorVS: Person Identification via Video Trajectory-Sensor Correspondence in a Real-World Warehouse 査読有り Open Access

Kano K., Mori Y., Katayama S., Urano K., Yonezawa T., Kawaguchi N.

Proceedings of the 15th International Conference on Indoor Positioning and Indoor Navigation Ipin 2025 2025年9月

Design and Deployment of a Private 5G Testbed Integrated with a Nationwide Academic Backbone 査読有り

Sasayama K., Nakao A., Takezawa H., Kawaguchi N., Maruyama M., Kondo T., Shimoji H., Urushidani S., Kurimoto T.

2025 IEEE Conference on Standards for Communications and Networking Cscn 2025 2025年9月

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：2025 IEEE Conference on Standards for Communications and Networking Cscn 2025

This paper presents a nationwide academic testbed that integrates private 5G mobile networks, known as "local 5G"in Japan, with SINET6, the country's ultra-high-speed and low-latency academic backbone network. Unlike public 5G networks, local 5G systems offer advanced features such as flexible uplink/downlink resource allocation and independent deployment of core and radio access functions. We developed a novel architecture in which both the 5G core and RAN are integrated into a compact, portable system connected to SINET via campus LANs. Deployed across multiple universities in Japan, this testbed enables high-performance mobile data communication tailored for academic research. We report on its design, implementation, performance evaluations, and a range of practical use case demonstrations conducted by participating institutions.

DOI： 10.1109/CSCN67557.2025.11230704

Scopus

Mitigating Latency Effects in VR180 Teleoperation via Virtual Robotic Arm Overlay 査読有り

Tsutsui S., Watanabe K., Usami R., Gushi Y., Watanabe Y., Shimura K., Hayashida N., Jimenez F., Urano K., Yonezawa T., Kawaguchi N.

15th International Conference on Mobile Computing and Ubiquitous Networking Icmu 2025 2025年9月

MetaUniTour: Enhancing Engagement and Intercultural Understanding through A Metaverse Tourism Platform 査読有り

Matsumoto K., Watanabe Y., Shimura K., Hayashida N., Katayama S., Urano K., Kawaguchi N., Aoki S., Leleito E., Yonezawa T.

Proceedings 2025 International Conference on Metaverse Computing Networking and Applications Metacom 2025 頁： 125 - 130 2025年8月

Discovering Emerging Urban Hotspots Using Satellite Imagery and Human Mobility Data 査読有り

Tillayeva N., Terashima H., Tamura N., Shoji K., Hossain T., Katayama S., Urano K., Yonezawa T., Kawaguchi N.

Distributed, Ambient and Pervasive Interactions: 13th International Conference, DAPI 2025, Held as Part of the 27th HCI International Conference, HCII 2025 15803 LNCS 巻頁： 110 - 120 2025年6月

Clustering-Based Multitask Classification for Predictive Consumer Behavior Modeling 査読有り国際共著

2025年4月

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：2025 International Conference on Activity and Behavior Computing Abc 2025

Understanding customers and segmenting in an appropriate way play a vital role in business operations and effective marketing strategies. It helps target specific customer groups for increased engagement and profit. However, traditional approaches often struggle to predict future purchase patterns, especially when user data is sparse and fragmented across multiple sources. This research proposes a novel clustering-based multitask classification method to address these limitations. We integrated the Recency, Frequency Monetary (RFM) model to cluster customers based on their purchase patterns from various purchase categories (tickets, merchandise, fan club membership) to create a more comprehensive understanding of loyalty for a professional basketball team in Japan. The proposed model predicts a customer’s segment across all purchase categories, implying the potential relationships between each category. Initial results demonstrate that the model effectively analyzes purchase behaviors, offering actionable insights to enhance marketing strategies.

DOI： 10.1109/ABC64332.2025.11118620

Web of Science

Scopus

Efficient Edge AI based Annotation and Detection Framework for Logistics Warehouses 査読有り

Yuki Mori, Yusuke Asai, Keisuke Higashiura, Shin Katayama, Kenta Urano, Takuro Yonezawa, Nobuo Kawaguchi

IEEE Consumer Communications & Networking Conference 2025 2025年1月

　前のページ - 次のページ　

<河口 信夫>

論文 - 河口 信夫

<河口　信夫>

論文 - 河口　信夫