陸 慧敏 (リク ケイビン)

LU Huimin

写真a

職名

准教授

研究室住所

福岡県北九州市戸畑区仙水町1-1

研究分野・キーワード

人工知能、ロボティックス、海中光学、コンピュータビジョン

ホームページ

http://www.ericlab.org

取得学位 【 表示 / 非表示

  • 九州工業大学 -  博士(工学)  2014年03月

学内職務経歴 【 表示 / 非表示

  • 2019年09月
    -
    継続中

    九州工業大学   大学院工学研究院   機械知能工学研究系   准教授  

所属学会・委員会 【 表示 / 非表示

  • 2020年04月
    -
    継続中
     

    電子情報通信学会  日本国

  • 2019年12月
    -
    継続中
     

    情報処理学会  日本国

  • 2019年08月
    -
    継続中
     

    SPIE  アメリカ合衆国

  • 2012年01月
    -
    継続中
     

    IEEE  アメリカ合衆国

専門分野(科研費分類) 【 表示 / 非表示

  • 知覚情報処理

  • 計測工学

 

論文 【 表示 / 非表示

  • Deep hierarchical encoding model for sentence semantic matching

    Lu W., Zhang X., Lu H., Li F.

    Journal of Visual Communication and Image Representation    71   2020年08月  [査読有り]

     概要を見る

    © 2020 Elsevier Inc. Sentence semantic matching (SSM) always plays a critical role in natural language processing. Measuring the intrinsic semantic similarity among sentences is very challenging and has not been substantially addressed. The latest SSM research usually relies on a shallow text representation and interaction between sentence pairs, which might not be enough to capture the complex semantic features and lead to limited performance. To capture more semantic context features and interactions, we propose a hierarchical encoding model (HEM) for sentence representation, further enhanced by a hierarchical matching mechanism for sentence interaction. Given two sentences, HEM generates intermediate and final representations in encoding layer, which are further handled by a novel hierarchical matching mechanism to capture more multi-view interactions in matching layer. The comprehensive experiments demonstrate that our model is capable to capture more sentence semantic features and interactions, which significantly outperforms the existing state-of-the-art neural models on the public real-world dataset.

    DOI Scopus

  • Dynamics and Isotropic Control of Parallel Mechanisms for Vibration Isolation

    Yang X., Wu H., Li Y., Kang S., Chen B., Lu H., Lee C.K.M., Ji P.

    IEEE/ASME Transactions on Mechatronics    25 ( 4 ) 2027 - 2034   2020年08月  [査読有り]

     概要を見る

    © 1996-2012 IEEE. Parallel mechanisms have been employed as architectures of high-precision vibration isolation systems. However, their performances in all degrees of freedom (DOFs) are nonidentical. The conventional solution to this problem is isotropic mechanism design, which is laborious and can hardly be achieved. This article proposes a novel concept; namely, isotropic control, to solve this problem. Dynamic equations of parallel mechanisms with base excitation are established and analyzed. An isotropic control framework is then synthesized in modal space. We derive an explicit relationship between modal control force and actuation force in joint space, enabling implementation of the isotropic controller. The multi-DOF system is transformed into multiidentical single-DOF systems. Under the framework of isotropic control, parallel mechanisms obtain an identical frequency response for all modes. An identical corner frequency, active damping, and rate of low-frequency transmissibility are achieved for all modes using a combining proportional, integral, and double integral compensator as a subcontroller. A 6-UPS parallel mechanism is presented as an example to demonstrate effectiveness of the new approach.

    DOI Scopus

  • Correlated Features Synthesis and Alignment for Zero-shot Cross-modal Retrieval

    Xu X., Lin K., Lu H., Gao L., Shen H.T.

    SIGIR 2020 - Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval      1419 - 1428   2020年07月  [査読有り]

     概要を見る

    © 2020 ACM. The goal of cross-modal retrieval is to search for semantically similar instances in one modality by using a query from another modality. Existing approaches mainly consider the standard scenario that requires the source set for training and the target set for testing share the same scope of classes. However, they may not generalize well on zero-shot cross-modal retrieval (ZS-CMR) task, where the target set contains unseen classes that are disjoint with the seen classes in the source set. This task is more challenging due to 1) the absence of the unseen classes during training, 2) inconsistent semantics across seen and unseen classes, and 3) the heterogeneous multimodal distributions between the source and target set. To address these issues, we propose a novel Correlated Feature Synthesis and Alignment (CFSA) approach to integrate multimodal feature synthesis, common space learning and knowledge transfer for ZS-CMR. Our CFSA first utilizes class-level word embeddings to guide two coupled Wassertein generative adversarial networks (WGANs) to synthesize sufficient multimodal features with semantic correlation for stable training. Then the synthetic and true multimodal features are jointly mapped to a common semantic space via an effective distribution alignment scheme, where the cross-modal correlations of different semantic features are captured and the knowledge can be transferred to the unseen classes under the cycle-consistency constraint. Experiments on four benchmark datasets for image-text retrieval and two large-scale datasets for image-sketch retrieval show the remarkable improvements achieved by our CFAS method comparing with a bundle of state-of-the-art approaches.

    DOI Scopus

  • Data Analytics for the COVID-19 Epidemic

    Wang R., Hu G., Jiang C., Lu H., Zhang Y.

    Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020      1261 - 1266   2020年07月  [査読有り]

     概要を見る

    © 2020 IEEE. With the spread of COVID-19 worldwide, people¡¯s production and life have been significantly affected. Artificial intelligence and big data technologies have been vigorously developed in recent years. It is very significant to use data science and technology to help humans in a timely and accurate manner to prevent and control the development of the epidemic, maintain social stability and assess the impact of the epidemic. This paper explores how data science can play a role from the perspectives of epidemiology, social networking, and economics. In particular, for the existing epidemic model SIR, we present a parameter learning method using particle swarm optimization (PSO) and the least squares method, and use it to predict the trend of the epidemic. Aiming at the social network data, we provide a specific method to realize sentiment analysis during the epidemic and propose an explainable fake news detection technique based on a variety of data mining methods.

    DOI Scopus

  • Deep Learning for Visual Segmentation: A Review

    Sun J., Li Y., Lu H., Kamiya T., Serikawa S.

    Proceedings - 2020 IEEE 44th Annual Computers, Software, and Applications Conference, COMPSAC 2020      1256 - 1260   2020年07月  [査読有り]

     概要を見る

    © 2020 IEEE. Big data-driven deep learning methods have been widely used in image or video segmentation. The main challenge is that a large amount of labeled data is required in training deep learning models, which is important in real-world applications. To the best of our knowledge, there exist few researches in the deep learning-based visual segmentation. To this end, this paper summarizes the algorithms and current situation of image or video segmentation technologies based on deep learning and point out the future trends. The characteristics of segmentation that based on semi-supervised or unsupervised learning, all of the recent novel methods are summarized in this paper. The principle, advantages and disadvantages of each algorithms are also compared and analyzed.

    DOI Scopus

全件表示 >>

著書 【 表示 / 非表示

  • Cognitive Internet of Things: Frameworks, Tools and Applications

    Huimin LU ( 単著 )

    Springer International Publishing  2020年01月 ISBN: 978-3-030-04945-4

  • Artificial Inteligence and Robotics

    Huimin Lu, Xing Xu ( 共編者 )

    Springer  2017年12月 ISBN: 978-3-319-69877-9

  • Artificial Intelligence and Computer Vision

    Huimin Lu, Yujie Li ( 共編者 )

    Springer  2016年11月

講演 【 表示 / 非表示

  • AIを活用した水中画像処理技術と深海資源調査への展開

    第1回海中海底工学フォーラム・ZERO   2019年04月12日 

  • Artificial Intelligence in Deep-sea Observing

    The 2nd International Symposium on Artificial Intelligence and Robotics 2017 ( Kitakyushu, Japan )  2017年11月25日  ISAIR

  • Extreme Optical Imaging for Deep-sea Observing Network

    26th International Electrotechnical and Computer Science Conference ERK 2017 ( Congress Center Bernardin, Portorož, Slovenia )  2017年09月25日  IEEE Slovenia

  • Next Generation Artificial Intelligence in Society 5.0

    IBM Australia Seminar ( Melbourne, Australia )  2017年09月04日  IBM Australia

科研費獲得実績 【 表示 / 非表示

  • 日中超スマート社会の実現に向けた次世代のAI/IoTに関する研究

    二国間国際交流事業

    研究期間:  2018年04月  -  2019年03月

    研究課題番号:  00000001

  • 深海採鉱機採削時の画像計測システムの研究開発

    若手研究(B)

    研究期間:  2017年04月  -  2019年03月

    研究課題番号:  17K14694

  • 深海採鉱機向けリアルタイム小型イメージングシステムの研究開発

    特別研究員奨励費

    研究期間:  2015年04月  -  2016年09月

    研究課題番号:  15F15077

  • 深海採鉱機向け鉱床計測用リアルタイム画像採取処理装置の研究開発

    特別研究員奨励費

    研究期間:  2013年04月  -  2015年03月

    研究課題番号:  13J10713

受託研究・共同研究実施実績 【 表示 / 非表示

  • 国立研究開発法人情報通信研究機構国際交流プログラム

    受託研究

    研究期間:  2018年04月  -  2019年03月

  • 国立情報学研究所共同研究

    受託研究

    研究期間:  2018年04月  -  2019年03月

寄附金・講座 【 表示 / 非表示

  • 公益財団法人電気通信普及財団 研究調査助成

    公益財団法人電気通信普及財団  2019年05月

  • 電気通信普及財団研究調査助成

    公益財団法人電気通信普及財団  2018年05月

  • 造船学術研究推進機構 助成金

    造船学術研究推進機構  2017年08月

  • 電気通信普及財団 研究調査助成

    公益財団法人電気通信普及財団  2017年04月

 

学会・委員会等活動 【 表示 / 非表示

  • 2019年08月
    -
    継続中

    IEEE Computer Society Big Data Special Technical Committee   共同委員長

 

国際会議の開催 【 表示 / 非表示

  • EAI International Conference on Robotic Sensor Networks

    2017年11月25日  -  2017年11月26日 

  • The 2nd International Symposium on Artificial Intelligence and Robotics 2017

    2017年11月25日  -  2017年11月26日 

  • The 1st International Symposium on Artificial Intelligence and Robotics 2016

    China  2016年12月13日  -  2016年12月13日  Huimin Lu

国際交流窓口担当 【 表示 / 非表示

  • リュブリャナ大学  2018年11月  -  継続中

  • 南京郵電大学 オートメーション工学部  2018年05月  -  継続中