KAMEYAMA Keisuke

Affiliation
Institute of Systems and Information Engineering
Official title
Professor
URL
Email
=l(,68.(Ol$0(<$0$FUQ&6O768.8%$O$&O-3\.$0(<$0$O.(,68.(O*7FUQ8O768.8%$O$&O-3
Research fields
Intelligent informatics
Research keywords
neural network
pattern recognition
learning
adaptive information processing
image retrieval
Research projects
Model Switching Criteria in Dynamic Model Learning of Neural Networks2019-04 -- 2023-03KAMEYAMA KeisukeJapan Society for the Promotion of Science/Grant-in-Aid for Scientific Research(C)4,420,000Yen
Signal Recognition Mechanisms by Selecting Higher-Order Spectral Features Through Learning2016-04 -- 2020-03KAMEYAMA KeisukeJapan Society for the Promotion of Science/Grant-in-Aid for Scientific Research(C)4,550,000Yen
Statistical pattern recognition -- (current)/
Content-based multimedia database -- (current)/
Pattern recognition and signal processing using higher-order statistics -- (current)/
Learning in neural networks -- (current)/
Simultaneous search of model and parameters in pattern recognition systems. -- (current)/
Framework of Novel Image Feature Extraction using Higher-Order Statistical Kernel Functions2010 -- 2012KAMEYAMA KeisukeJapan Society for the Promotion of Science/Grant-in-Aid for Scientific Research (C)4,290,000Yen
Meta-Learning Algorithms in Neural Networks and their Implementation in Parallel Distributed Systems2003 -- 2005KAMEYAMA KeisukeJapan Society for the Promotion of Science/Grant-in-Aid for Young Scientists (B)2,900,000Yen
Search of Local Interactions for Improved Generalization in Neural Network Learning2000 -- 2001Japan Society for the Promotion of Science/Grant-in-Aid for Encouragement of Scientists (A)2,500,000Yen
Career history
2000-02 -- 2007-01University of TsukubaAssistant Professor
2007-02 -- 2014-02University of TsukubaAssociate Professor
2014-03 -- (current)University of TsukubaFaculty of Engineering, Information and SystemsProfessor
Academic background
-- 1989Tokyo Institute of Technology Faculty of Engineering Department of Electric and Electronic Engineering
-- 1991-03Tokyo Institute of Technology Graduate School, Division of Integrated Science and Engineering Precision Machinery Systems
Degree
1999-09PhD(Engineering)Tokyo Institute of Technology
Licenses and qualifications
2009-07-29On-the-Ground Services II- Category Special Radio Operator
Academic societies
1991 -- (current)Institute of Electrical and Electronics Engineers (IEEE)
1990 -- (current)The Institute of Electronics, Information and Communication Engineers.
2010 -- (current)Japanese Neural Network Society (JNNS)
Articles
  • Affinity-Weighted RandAugment for Problem-Oriented Augmentation in Image Classification
    PARK Yuna; TAKASE Tomoumi; KAMEYAMA Keisuke; ONISHI M...
    Proceedings of International Joint Conference on Neural Networks 2024/pp.1-8, 2024-06
  • PINN with Antisymmetric RNN for Solving Nonlinear Partial Differential Equations
    Maniamfu Pavodi Ndoyi; Ali U. A. Md. Ehsan; Kameyama K...
    International Conference on Scientific Computation and Machine Learning 2024, 2024-03
  • CNN Model Compression by Merit-Based Distillation
    Morikawa Takumi; Kameyama Keisuke
    2023 19th IEEE International Colloquium on Signal Processing & Its Applications (CSPA)/pp.122-127, 2023-03
  • Machine Learning Curriculums Generated by Classifier Ensembles
    Huang Tzu-Jui; Kameyama Keisuke
    2023 19th IEEE International Colloquium on Signal Processing & Its Applications (CSPA)/pp.117-121, 2023-03
  • LSTM-based Forecasting using Policy Stringency and Time-varying Parameters of the SIR Model for COVID-19
    Maniamfu Pavodi Ndoyi; Kameyama Keisuke
    2023 19th IEEE International Colloquium on Signal Processing & Its Applications (CSPA)/pp.111-116, 2023-03
  • Informative Band Subset Selection for Hyperspectral Image Classification using Joint and Conditional Mutual Information
    Ali U. A. Md Ehsan; Kameyama Keisuke
    IEEE Symposium on Computational Intelligence in Remote Sensing (IEEE SSCI 2022)/pp.573-580, 2022-12
  • Joint-Conditional Mutual Information based Feature Subset Selection for Remotely Sensed Hyperspectral Image Classification
    Ali U. A. Md Ehsan; Kameyama Keisuke
    Technical Report of IEICE/122(90)/pp.115-122, 2022-06
  • Iris Authentication using Higher-Order Spectral Features
    亀山 啓輔
    Technical Report of IEICE/121(231)/pp.31-35, 2021-11
  • Multi-Stage Model Compression using Teacher Assistant and Distillation with Hint-Based Training
    Morikawa Takumi; Kameyama Keisuke
    2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops)/pp.484-490, 2022-03
  • Adaptive Selection of Classifiers for Person Recognition by Iris Pattern and Periocular Image
    Ogawa Keita; Kameyama Keisuke
    Neural Information Processing. ICONIP 2021. Lecture Notes in Computer Science/13111/pp.656-667, 2021-12
  • Pre-Training Acquisition Functions by Deep Reinforcement Learning for Fixed Budget Active Learning
    Taguchi Yusuke; Hino Hideitsu; Kameyama Keisuke
    Neural Processing Letters/53/pp.1945-1962, 2021-02
  • Metal Material Analysis using SEM Images of Multiple Magnifications
    伊藤 龍之介; 亀山 啓輔; 日野 英逸
    電子情報通信学会技術研究報告 (信学技報)/119(481)/pp.71-76, 2020-03
  • Active learning for multiple data selection using reinforcement learning
    田口 優介; 日野 英逸; 亀山 啓輔
    Technical Report of IEICE/119(360)/pp.99-106, 2020-01
  • Recognition feature prediction from low-resolution iris images using CNN
    渡邉 崚; 亀山 啓輔
    Technical Report of IEICE/119(214)/pp.5-10, 2019-10
  • Active Learning with Interpretable Predictor
    Taguchi Yusuke; Kameyama Keisuke; Hino Hideitsu
    Proceedings of International Joint Conference on Neural Networks (IJCNN)/pp.1-8, 2019-07
  • Parallel Cooperative Ensemble Learning by Adaptive Data Weighting and Error-Correcting Output Codes
    Utsumi Shota; Kameyama Keisuke
    Lecture Notes in Computer Science (Proc. ICONIP 2018)/11303/pp.673-683, 2018-12
  • Revising the Algorithm of Ensenble Learning by an Index of Complementarity among Weak Learners
    内海 翔太; 亀山 啓輔
    Technical Report of IEICE/118(284, IBISML2018-102)/pp.429-434, 2018-11
  • Parameter Density Inheritance Using Kernel Density Estimation for Efficient CNN Learning
    堀内 圭佑; 亀山 啓輔
    Technical Report of IEICE/118(284, IBISML2018-45)/pp.9-15, 2018-11
  • Interpretable active sampling
    田口 優介; 亀山 啓輔; 日野 英逸
    Technical Report of IEICE/118(220, IBISML2018-27)/pp.97-104, 2018-09
  • Arrangement of Complementary Weak Learners using Weights Assigned to Data in Parallel Ensemble Learning
    内海 翔太; 亀山 啓輔
    Technical Report of IEICE/118(220)/pp.9-15, 2018-09
  • Parameter Density Inheritance Using Kernel Density Estimation for Efficient CNN Learning
    Horiuchi Keisuke; Kameyama Keisuke
    Proceedings of 2018 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)/pp.308-313, 2018-12
  • Automatic MOOC video classification using transcript features and convolutional neural networks
    Chatbri Houssem; McGuinness Kevin; Little Suzanne; Zhou ...
    ACM Multimedia 2017 - MultiEdTech Workshop/pp.21-26, 2017-10
  • Lapped cuboid-based perceptual encryption for Motion JPEG standard
    Shimizu Kosuke; Suzuki Taizo; Kameyama Keisuke
    Proc. APSIPA ASC 2018/pp.2022-2026, 2018-11
  • Cube-based encryption-then-compression system for video sequences
    Shimizu Kosuke; Suzuki Taizo; Kameyama Keisuke
    IEICE Transactions on Fundamentals of Electronics, Communictions and Computer Sciences/E101-A(11)/pp.1815-1822, 2018-11
    CiNii
  • Log analysis for hijacked user account's login detection method based on an autonomous system number of access source
    Sato Akira; Sannomiya Shuji; Oyama Yoshihiro; Yoshida Ke...
    信学技報IA/117(299)/pp.95-98, 2017-11
  • more...
Books
  • Model Switching Maneuvers in Pattern Recognition Systems
    Kameyama Keisuke
    1999-09
  • リスク工学の視点とアプローチ ―現代生活に潜むリスクにどう取り組むかー
    古川宏; 佐藤美佳; 亀山啓輔; 谷口綾子; 梅本通孝; 羽田野祐子
    Corona-sha, 2009-01
Conference, etc.
  • 教師モデルの推論時の着目領域情報を使用したDistillationによるCNNのモデル圧縮
    白浜拓海; 亀山 啓輔
    2024年電子情報通信学会総合大会/2024-03-04--2024-03-08
  • 事前学習を用いたクラスタリングによるログデータの異常検知
    渡辺耕太朗; 亀山 啓輔
    2024年電子情報通信学会総合大会/2024-03-04--2024-03-08
  • CNNに対するフィルタ刈り込みと欠損した情報の補償による性能低下の抑制
    亀山 啓輔
    第22回情報科学技術フォーラム(FIT)/2023-09-06--2023-09-08
  • Data Augmentation using Adversarial Sample Generation
    朴 潤花; 亀山 啓輔
    IEICE General Conference 2023/2023-03-07--2023-03-10
  • Lip Video-Assisted Speech Recognition by Feature Exchange in 3DCNN and MLP
    河内 信誠; 亀山 啓輔
    IEICE General Conference 2023/2023-03-07--2023-03-10
  • Neural Network Model Compression by using Teacher Assistant and Distillation with Hint Learning
    森川 拓海; 亀山 啓輔
    Forum on Information Technology 2021/2021/8/25--2021/8/27
  • Multimodal personal authentication using iris and periocular information by learning combination of classifiers
    小川 恵太; 亀山 啓輔
    Forum on Information Technology 2021/2021/8/25--2021/8/27
  • キャンパスネットワークにおけるインシデント対応の自動化– シミュレーション実験の報告 –
    佐藤 聡; 三宮 秀次; 片岸 一起; 中井 央; 亀山 啓輔
    大学ICT推進協議会 2019年度　年次大会/2019-12-12--2019-12-14
  • JPEG perceptual encryption with bit-pair permutation in quantized DCT coefficients
    淸水 恒輔; 鈴木 大三; 亀山 啓輔
    IEICE 33rd Signal Processing Symposium/2018-11-06--2018-11-08
  • Reversible data hiding for two-layer HDR image coding using difference expansion into residual signals
    楊 凱; 淸水 恒輔; 鈴木 大三; 亀山 啓輔
    IEICE 33rd Signal Processing Symposium/2018-11-06--2018-11-08
  • 3-D human pose estimation from 2-D still image by least-square regression using convolutional neural network
    長谷川 輝; 鈴木 大三; 亀山 啓輔
    IEICE General Conference/2017-03-22--2017-03-25
  • Human hand detection based on skin color region detection and HOG features
    堀内 圭佑; 鈴木 大三; 亀山 啓輔
    IEICE General Conference/2017-03-22--2017-03-25
  • Palm vein authentication using shape context descriptor
    杉原 伊織; 鈴木 大三; 亀山 啓輔
    IEICE General Conference/2017-03-22--2017-03-25
  • Music recommendation by learning the user's association between music and color
    佐久間 一生; 鈴木 大三; 亀山 啓輔
    IEICE General Conference/2017-03-22--2017-03-25
  • A Comparative Study of Keypoint Extraction Methods in Binary Image Matching
    Houssem Chatbri; Keisuke Kameyama
    Tunisia-Japan Symposium on Science, Society and Technology (TJASSST 2016)/2016-02
  • Partial i mage retrieval based on color feature invariant to specific illumination variation
    澤 明香利; 松本 亘; 亀山 啓輔
    IEICE General Conference 2016/2015-03-10--2015-03-13
  • Adaptive Selection of Feature Extraction Region in Iris Authentication using Local Binary Pattern
    相澤 美晴; 亀山 啓輔
    IEICE General Conference 2016/2015-03-10--2015-03-13
  • Construction of Braille Recognition System Using a Smartphone
    蓑田 豪; 亀山 啓輔
    IEICE General Conference 2016/2015-03-10--2015-03-13
  • Hand Detection Using One-Class SVM
    Haryadi Harry; Kameyama Keisuke
    IEICE General Conference 2016/2015-03-10--2015-03-13
  • Optical Character Re cognition robust to the difference in font using Support Region Descriptor
    安藤 洸野; Chatbri Houssem; 亀山 啓輔
    IEICE General Conference 2016/2015-03-10--2015-03-13
  • Thinning Noisy Sketch Images Using Scale Space Filtering
    Chatbri Houssem; Kameyama Keisuke
    10th AEARU Workshop on Computer Science and Web Technology/2015-02-25--2015-02-27
  • Improvement of Particle Swarm Optimization Using Pseudo-Gradient Information
    Utsumi Shota; Kameyama Keisuke
    10th AEARU Workshop on Computer Science and Web Technology/2015-02-25--2015-02-27
  • Adaptive image interpolation method using a local gradient feature value
    川崎 政吾; 亀山 啓輔
    第13回情報科学技術フォーラム(FIT 2014)/2014-09-03--2014-09-05
  • Improvement of Particle Swarm Optimization Using Pseudo-Gradient Information
    内海 翔太; 亀山 啓輔
    第13回情報科学技術フォーラム(FIT 2014)/2014-09-03--2014-09-05
  • Content-Based Image Retrieval Using Adaptive Thinning and Illumination Invariant Color Features
    Kameyama Keisuke; Chatbri Houssem; Matsumoto Wataru
    Tunisia-Japan Symposium on Science, Society and Technology (TJASSST2013)/2013-12
  • more...
Intellectural property rights
  • 逆問題解析ネットワークおよびその演算方法
    小杉幸夫; 亀山啓輔
  • 地理画像変化域の抽出方法，及び地理画像変化域の抽出装置
    小杉幸夫; 亀山啓輔; 高石秀樹; 森谷直哉; 角本繁; 土居原健
  • 画像処理装置，方法およびプログラム
    寅市和男; 亀山啓輔; 平野華世子
Teaching
2024-10 -- 2025-02Computer Science Seminar AfUniversity of Tsukuba.
2024-10 -- 2025-02Computer Science Seminar BfUniversity of Tsukuba.
2024-04 -- 2024-08Research in Computer Science IsUniversity of Tsukuba.
2024-10 -- 2025-02Research in Computer Science IIfUniversity of Tsukuba.
2024-04 -- 2024-08Computer Science Seminar BUniversity of Tsukuba.
2024-10 -- 2025-02Computer Science Seminar BUniversity of Tsukuba.
2024-10 -- 2025-02Research in Computer Science DUniversity of Tsukuba.
2024-10 -- 2025-02Research in Computer Science BUniversity of Tsukuba.
2024-10 -- 2025-02Research in Computer Science fUniversity of Tsukuba.
2024-10 -- 2025-02Research in Computer Science AUniversity of Tsukuba.
more...
Professional activities
2022-07 -- (current)NIPC監事
2020-07 -- 2021-06NIPCGeneral Chair
2018-04 -- 2018-09International Conference on Neural Information Processing (ICONIP 2018)Program Commitee Member
2017-04 -- 2017-09International Conference on Neural Information Processing (ICONIP 2017)Program Commitee Member
2011-05 -- (current)The Institute of Electronics, Information and Communication Enineers (IEICE)ソサイエティ論文誌編集委員 査読委員
2013-03 -- 2013-11International Conference on Neural Information Processing (ICONIP 2013)Program Committee Member
University Management
2021-04 -- (current)Organization for Information InfrastructureDirector
2016-04 -- 2022-03学術情報メディアセンターセンター長

(Last updated: 2024-09-27)