Updated on 2024/08/22

写真a

 
YOSHIDA,Soh
 
Organization
Faculty of Engineering Science Associate Professor
Title
Associate Professor
External link

Degree

  • 博士(情報科学)(北海道大学) ( 2016.3 )

Research Interests

  • ソーシャルメディア

  • Webマイニング

  • マルチメディア処理

  • Information retrieval

Research Areas

  • Informatics / Intelligent informatics

  • Informatics / Web informatics and service informatics

Education

  • Hokkaido University   Graduate School, Division of Information Science

    2014.4 - 2016.3

      More details

  • Hokkaido University   Graduate School, Division of Information Science

    2012.4 - 2014.3

      More details

  • Hokkaido University   Faculty of Engineering   Department of Electronics and Information Engineering

    2008.4 - 2012.3

      More details

Research History

  • Kansai University   Faculty of Engineering Science   Associate Professor

    2023.4

      More details

    Country:Japan

    researchmap

  • Kansai University   Faculty of Engineering Science   Assistant Professor

    2016.4 - 2023.3

      More details

Professional Memberships

  • 電子情報通信学会

      More details

  • 映像情報メディア学会

      More details

  • ACM(Association for Computing Machinery)

      More details

  • IEEE(The Institute of Electrical and Electronics Engineers, Inc.)

      More details

Committee Memberships

  • スマートインフォメディアシステム研究専門委員会   専門委員  

    2024.6 - Present   

      More details

  • スマートインフォメディアシステム研究専門委員会   幹事  

    2022.6 - 2024.6   

      More details

  • 電子情報通信学会   英文論文誌小特集編集委員会 編集幹事  

    2021.12 - 2024.3   

      More details

    Committee type:Academic society

    researchmap

  • 電子情報通信学会 英文論文誌A編集委員会   編集委員  

    2020.6 - 2024.6   

      More details

    Committee type:Academic society

    researchmap

  • スマートインフォメディアシステム研究専門委員会   幹事補佐  

    2020.6 - 2022.6   

      More details

    Committee type:Academic society

    researchmap

  • スマートインフォメディアシステム研究専門委員会   専門委員  

    2019.6 - 2020.5   

      More details

    Committee type:Academic society

    researchmap

  • ディジタル信号処理システム最適化技術調査専門委員会   専門委員  

    2017.6 - 2020.5   

      More details

    Committee type:Academic society

    researchmap

▼display all

Papers

  • Graph-Based Interpretability for Fake News Detection through Topic- and Propagation-Aware Visualization Reviewed

    Kayato Soga, Soh Yoshida, Mitsuji Muneyasu

    Computation   12 ( 4 )   82 - 82   2024.4

     More details

    Authorship:Corresponding author   Language:English   Publishing type:Research paper (scientific journal)   Publisher:MDPI AG  

    In the context of the increasing spread of misinformation via social network services, in this study, we addressed the critical challenge of detecting and explaining the spread of fake news. Early detection methods focused on content analysis, whereas recent approaches have exploited the distinctive propagation patterns of fake news to analyze network graphs of news sharing. However, these accurate methods lack accountability and provide little insight into the reasoning behind their classifications. We aimed to fill this gap by elucidating the structural differences in the spread of fake and real news, with a focus on opinion consensus within these structures. We present a novel method that improves the interpretability of graph-based propagation detectors by visualizing article topics and propagation structures using BERTopic for topic classification and analyzing the effect of topic agreement on propagation patterns. By applying this method to a real-world dataset and conducting a comprehensive case study, we not only demonstrated the effectiveness of the method in identifying characteristic propagation paths but also propose new metrics for evaluating the interpretability of the detection methods. Our results provide valuable insights into the structural behavior and patterns of news propagation, contributing to the development of more transparent and explainable fake news detection systems.

    DOI: 10.3390/computation12040082

    researchmap

  • ConfidentMix: Confidence-Guided Mixup for Learning With Noisy Labels Reviewed

    Ryota Higashimoto, Soh Yoshida, Mitsuji Muneyasu

    IEEE Access   12   58519 - 58531   2024.4

     More details

    Authorship:Corresponding author   Language:English   Publishing type:Research paper (scientific journal)   Publisher:Institute of Electrical and Electronics Engineers (IEEE)  

    DOI: 10.1109/access.2024.3393440

    researchmap

  • CRAS: Curriculum Regularization and Adaptive Semi-Supervised Learning with Noisy Labels Reviewed

    Ryota Higashimoto, Soh Yoshida, Mitsuji Muneyasu

    Applied Sciences   14 ( 3 )   1208 - 1208   2024.1

     More details

    Authorship:Corresponding author   Publishing type:Research paper (scientific journal)   Publisher:MDPI AG  

    This paper addresses the performance degradation of deep neural networks caused by learning with noisy labels. Recent research on this topic has exploited the memorization effect: networks fit data with clean labels during the early stages of learning and eventually memorize data with noisy labels. This property allows for the separation of clean and noisy samples from a loss distribution. In recent years, semi-supervised learning, which divides training data into a set of labeled clean samples and a set of unlabeled noisy samples, has achieved impressive results. However, this strategy has two significant problems: (1) the accuracy of dividing the data into clean and noisy samples depends strongly on the network’s performance, and (2) if the divided data are biased towards the unlabeled samples, there are few labeled samples, causing the network to overfit to the labels and leading to a poor generalization performance. To solve these problems, we propose the curriculum regularization and adaptive semi-supervised learning (CRAS) method. Its key ideas are (1) to train the network with robust regularization techniques as a warm-up before dividing the data, and (2) to control the strength of the regularization using loss weights that adaptively respond to data bias, which varies with each split at each training epoch. We evaluated the performance of CRAS on benchmark image classification datasets, CIFAR-10 and CIFAR-100, and real-world datasets, mini-WebVision and Clothing1M. The findings demonstrate that CRAS excels in handling noisy labels, resulting in a superior generalization and robustness to a range of noise rates, compared with the existing method.

    DOI: 10.3390/app14031208

    researchmap

  • Unbiased Pseudo-Labeling for Learning with Noisy Labels Reviewed

    Ryota HIGASHIMOTO, Soh YOSHIDA, Takashi HORIHATA, Mitsuji MUNEYASU

    IEICE Transactions on Information and Systems   E107.D ( 1 )   44 - 48   2024.1

     More details

    Authorship:Corresponding author   Publishing type:Research paper (scientific journal)   Publisher:Institute of Electronics, Information and Communications Engineers (IEICE)  

    DOI: 10.1587/transinf.2023mul0002

    researchmap

  • Exploiting stance similarity and graph neural networks for fake news detection Reviewed

    Kayato Soga, Soh Yoshida, Mitsuji Muneyasu

    Pattern Recognition Letters   177   26 - 32   2024.1

     More details

    Authorship:Corresponding author   Language:English   Publishing type:Research paper (scientific journal)   Publisher:Elsevier BV  

    DOI: 10.1016/j.patrec.2023.11.019

    researchmap

  • U-Net Architecture for Ancient Handwritten Chinese Character Detection in Han Dynasty Wooden Slips Reviewed

    Hojun SHIMOYAMA, Soh YOSHIDA, Takao FUJITA, Mitsuji MUNEYASU

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E106.A ( 11 )   1406 - 1415   2023.11

     More details

    Authorship:Corresponding author   Language:English   Publishing type:Research paper (scientific journal)   Publisher:Institute of Electronics, Information and Communications Engineers (IEICE)  

    DOI: 10.1587/transfun.2023smp0007

    researchmap

  • New Performance Evaluation Method for Data Embedding Techniques for Printed Images Using Mobile Devices Based on a GAN Reviewed

    Masahiro YASUDA, Soh YOSHIDA, Mitsuji MUNEYASU

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E106.A ( 3 )   481 - 485   2023.3

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:Institute of Electronics, Information and Communications Engineers (IEICE)  

    DOI: 10.1587/transfun.2022sml0003

    researchmap

  • Training Robust Deep Neural Networks on Noisy Labels Using Adaptive Sample Selection with Disagreement Reviewed

    Hiroshi Takeda, Soh Yoshida, Mitsuji Muneyasu

    IEEE Access   9   141131 - 141143   2021.10

     More details

    Authorship:Corresponding author   Language:English   Publishing type:Research paper (scientific journal)   Publisher:Institute of Electrical and Electronics Engineers (IEEE)  

    DOI: 10.1109/access.2021.3119582

    researchmap

  • Heterogeneous-Graph-Based Video Search Reranking Using Topic Relevance. Reviewed

    Soh Yoshida, Mitsuji Muneyasu, Takahiro Ogawa, Miki Haseyama

    IEICE Trans. Fundam. Electron. Commun. Comput. Sci.   Vol.E103-A ( No.12 )   1529 - 1540   2020.12

     More details

    Authorship:Lead author   Publishing type:Research paper (scientific journal)  

    DOI: 10.1587/transfun.2020SMP0023

    researchmap

    Other Link: https://dblp.uni-trier.de/db/journals/ieiceta/ieiceta103.html#YoshidaMOH20

  • Video Search Reranking with Relevance Feedback Using Visual and Textual Similarities Reviewed

    Takamasa FUJII, Soh YOSHIDA, Mitsuji MUNEYASU

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E102.A ( 12 )   1900 - 1909   2019.12

     More details

    Authorship:Corresponding author   Language:English   Publishing type:Research paper (scientific journal)   Publisher:Institute of Electronics, Information and Communications Engineers (IEICE)  

    DOI: 10.1587/transfun.e102.a.1900

    researchmap

  • Image Regularization with Total Variation and Optimized Morphological Gradient Priors Reviewed

    Shoya OOHARA, Mitsuji MUNEYASU, Soh YOSHIDA, Makoto NAKASHIZUKA

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   E102.A ( 12 )   1920 - 1924   2019.12

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:Institute of Electronics, Information and Communications Engineers (IEICE)  

    DOI: 10.1587/transfun.e102.a.1920

    researchmap

  • Graph-Based Video Search Reranking with Local and Global Consistency Analysis Reviewed

    Soh Yoshida, Takahiro Ogawa, Miki Haseyama, Mitsuji Muneyasu

    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS   E101D ( 5 )   1430 - 1440   2018.5

     More details

    Authorship:Lead author   Language:English   Publishing type:Research paper (scientific journal)   Publisher:IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG  

    Video reranking is an effective way for improving the retrieval performance of text-based video search engines. This paper proposes a graph-based Web video search reranking method with local and global consistency analysis. Generally, the graph-based reranking approach constructs a graph whose nodes and edges respectively correspond to videos and their pairwise similarities. A lot of reranking methods are built based on a scheme which regularizes the smoothness of pairwise relevance scores between adjacent nodes with regard to a user's query. However, since the overall consistency is measured by aggregating only the local consistency over each pair, errors in score estimation increase when noisy samples are included within query-relevant videos' neighbors. To deal with the noisy samples, the proposed method leverages the global consistency of the graph structure, which is different from the conventional methods. Specifically, in order to detect this consistency, the propose method introduces a spectral clustering algorithm which can detect video groups, in which videos have strong semantic correlation, on the graph. Furthermore, a new regularization term, which smooths ranking scores within the same group, is introduced to the reranking framework. Since the score regularization is performed by both local and global aspects simultaneously, the accurate score estimation becomes feasible. Experimental results obtained by applying the proposed method to a real-world video collection show its effectiveness.

    DOI: 10.1587/transinf.2017EDP7277

    Web of Science

    researchmap

  • Data Extraction Method from Printed Images with Different Formats Reviewed

    MUNEYASU Mitsuji, JINDA Nayuta, MORITANI Yuuya, YOSHIDA Soh

    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences   vol. E100-A, no.11, pp. 2355-2 ( 11 )   2355 - 2357   2017

     More details

    Language:English   Publisher:The Institute of Electronics, Information and Communication Engineers  

    <p>In this paper, we propose a method of embedding and detecting data in printed images with several formats, such as different resolutions and numbers of blocks, using the camera of a tablet device. To specify the resolution of an image and the number of blocks, invisible markers that are embedded in the amplitude domain of the discrete Fourier transform of the target image are used. The proposed method can increase the variety of images suitable for data embedding.</p>

    DOI: 10.1587/transfun.E100.A.2355

    researchmap

  • 歌謡番組における映像の構造に注目したシーン分割手法 Reviewed

    電子情報通信学会論文誌 D   J97-D ( 7 )   1177 - 1188   2014.7

     More details

    Authorship:Lead author   Language:Japanese   Publishing type:Research paper (scientific journal)  

    researchmap

  • 歌謡番組における映像の構造に注目したシーン分割手法 Reviewed

    吉田 壮

    電子情報通信学会論文誌(D)   vol.J97-D, no.7, pp. 1177-1188   2014.7

     More details

  • A method for improving SVM-Based image classification performance based on a target object detection scheme Reviewed

    Soh Yoshida, Hiroshi Okada, Takahiro Ogawa, Miki Haseyama

    ITE Transactions on Media Technology and Applications   1 ( 3 )   237 - 243   2013

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:Institute of Image Information and Television Engineers  

    This paper presents a new method to improve performance of SVM-based classification, which contains a target object detection scheme. The proposed method tries to detect target objects from training images and improve the performance of the image classification by calculating the hyperplane from the detection results. Specifically, the proposed method calculates a Support Vector Machine (SVM) hyperplane, and detects rectangular areas surrounding the target objects based on the distances between their feature vectors and the separating hyperplane in the feature space. Then modification of feature vectors becomes feasible by removing features that exist only in background areas. Furthermore, a new hyperplane is calculated by using the modified feature vectors. Since the removed features are not part of the target object, they are not relevant to the learning process. Therefore, their removal can improve the performance of the image classification. Experimental results obtained by applying the proposed methods to several existing SVM-based classification method show its effectiveness.

    DOI: 10.3169/mta.1.237

    Scopus

    researchmap

▼display all

MISC

  • Flexible Framework to Provide Explainability for Fake News Detection Methods on Social Media

    Hayato Matsumoto, Soh Yoshida, Mitsuji Muneyasu

    2022 IEEE 11th Global Conference on Consumer Electronics   2022.10

     More details

    Language:English  

    researchmap

  • A Robust Learning Framework Using Self-Supervised Learning for Learning With Noisy Labels

    Ryota Higashimoto, Soh Yoshida, Mitsuhu Muneyasu

    2022 IEEE 11th Global Conference on Consumer Electronics   2022.10

     More details

    Language:English  

    researchmap

  • Propagation-Based Fake News Detection Using a Combination of Different Content Features

    Kayato Soga, Soh Yoshida, Mitsuji Muneyasu

    2022 IEEE 11th Global Conference on Consumer Electronics   2022.10

     More details

    Language:English  

    researchmap

  • 学習初期の正則化と加重損失を用いたラベルノイズに頑健な半教師あり学習

    東本 良太, 吉田 壮, 棟安 実治

    IEICE スマートインフォメディアシステム研究会   2022.10

     More details

    Language:Japanese  

    researchmap

  • Method of Extracting Data from Images on a Curved Surface in Data Embedding to Printed Images Using Mobile Devices

    F. Kotegawa, M. Muneyasu, S. Yoshida

    2021.11

     More details

    Language:English  

    researchmap

  • Propagation-Based Fake News Detection Using Graph Neural Networks with Transformer

    H. Matsumoto, S. Yoshida, M. Muneyasu

    2021.10

     More details

    Language:English  

    researchmap

  • New Method of Detecting Calcification Regions in Dental Panoramic Radiographs Based on U-PraNet

    2021.10

     More details

    Language:English  

    researchmap

  • Detection of Calcification Regions from Dental Panoramic Radiographs Based on Semantic Segmentation Using Deep Learning

    T. Murano, M. Muneyasu, S. Yoshida, K. Chamnongthai, A. Asano, K. Uchida, N. Dewake, Y. Ishioka, N. Yoshinari

    2021 International Workshop on Smart Info-Media Systems in Asia   2021.9

     More details

    Language:English  

    researchmap

  • Method of Generating Pseudo-Captured Images to Evaluate the Performance of Data Embedding Techniques for Printed Images Using Mobile Devices

    M. Yasuda, M. Muneyasu, S. Yoshida

    2021.9

     More details

    Language:English  

    researchmap

  • 画素ごとに最適化した複数構造要素を用いたモルフォロジカル勾配に基づく画像の正則化

    岡 広高, 棟安 実治, 吉田 壮, 中静 真

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2020.10

     More details

  • Learning from Noisy Labeled Data Using Symmetric Cross-Entropy Loss for Image Classification

    Hiroshi Takeda, Soh Yoshida, Mitsuji Muneyasu

    2020 IEEE 9th Global Conference on Consumer Electronics   2020.10

     More details

  • 深層学習を用いた中国漢代木簡の文字領域検出

    吉田 壮, 谷知 真行, 藤田 高夫, 棟安 実治

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2020.3

     More details

  • 誤ラベルを含む教師データを用いたCNNによる画像認識

    武田 啓志, 吉田 壮, 棟安 実治

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2019.12

     More details

  • Data Retrieval from Printed Image Using Image Features and Data Embedding

    Takuhiro Nishikawa, Mitsuji Muneyasu, Yuuki Nishida, Soh Yoshida, Kosin Chamnongthai

    2019 International Symposium on Intelligent Signal Processing and Communication Systems   2019.12

     More details

  • Image Regularization with Morphological Gradient Priors Using Optimal Structuring Element for Each Pixel

    Shoya Oohara, Hirotaka Oka, Mitsuji Muneyasu, Soh Yoshida, Makoto Nakashizuka

    2019 International Symposium on Intelligent Signal Processing and Communication Systems   2019.12

     More details

  • 画像特徴量とデータ埋め込みを併用した印刷画像からのデータ取得

    西田 有希, 西川 拓宏, 棟安 実治, 吉田 壮

    第34回信号処理シンポジウム   2019.11

     More details

  • Tag-based Video Retrieval with Social Tag Relevance Learning

    Hiroshi Takeda, Soh Yoshida, Mitsuji Muneyasu

    2019 IEEE 8th Global Conference on Consumer Electronics   2019.10

     More details

  • 画素ごとの構造要素最適化を用いたモルフォロジカル勾配とTotal Variationに基づく画像の正則化

    大原 翔矢, 岡 広高, 棟安 実治, 吉田 壮, 中静 真

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2019.10

     More details

  • New Data Embedding and Detecting Method for Printed Image

    Masashi Shiiba, Mitsuji Muneyasu, Soh Yoshida

    2019 IEEE 8th Global Conference on Consumer Electronics   2019.10

     More details

  • 歯科パノラマX線写真における深層学習を用いた石灰化領域の検出精度向上

    山崎 康裕, 棟安 実治, 吉田 壮, 浅野 晃, 内田 啓一, 石岡 康明, 吉成 伸夫, 田口 明

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2019.10

     More details

  • 深層学習を用いた歯科パノラマX線写真における石灰化領域検出

    山崎 康裕, 棟安 実治, 吉田 壮, 浅野 晃, 内田 啓一, 石岡 康明, 吉成 伸夫, 田口 明

    電子情報通信学会2019年ソサイエティ大会   2019.9

     More details

  • A Graph-based Video Visual Reranking Method via Heterogenous Graph Analysis

    Soh Yoshida, Mitsuji Muneyasu

    2019 International Workshop on Smart Info-Media Systems in Asia   2019.9

     More details

  • データ埋め込みと画像特徴量を併用した印刷画像からのデータ取得

    西川 拓宏, 西田 有希, 棟安 実治, 吉田 壮

    電子情報通信学会2019年ソサイエティ大会   2019.9

     More details

  • Improvement of Data Embedding Method for Printed Image with High Detection Rate

    Masashi Shiiba, Takuhiro Nishikawa, Mitsuji Muneyasu, Soh Yoshida

    2019 International Workshop on Smart Info-Media Systems in Asia   2019.9

     More details

  • 印刷画像へのデータ埋め込みにおける検出手法の改善

    椎葉 将司, 棟安 実治, 吉田 壮

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2018.12

     More details

  • コンテンツの類似度に基づいた動画共有サイトで付与されたタグのランキング手法

    吉田 壮, 棟安 実治

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2018.12

     More details

  • Video Search Reranking on Multi-Layer Graphs Based on Combination of Video Features Using Subspace Analysis

    Soh Yoshida, Mitsuji Muneyasu

    2018 International Workshop on Smart Info-Media Systems in Asia   2018.12

     More details

  • マルチタグによるコンテンツ表現を考慮した映像のリランキング手法

    藤井 孝匡, 吉田 壮, 棟安 実治

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2018.12

     More details

  • Data Extraction for Data-embedded Printed Image Arranged at Arbitrary Positions

    Motoki Amami, Mitsuji Muneyasu, Soh Yoshida

    2018 International Workshop on Smart Info-Media Systems in Asia   2018.12

     More details

  • 印刷画像へのデータ埋め込みにおける検出率の改善法

    椎葉 将司, 棟安 実治, 吉田 壮

    第33回信号処理シンポジウム   2018.11

     More details

  • Image Regularization Using Total Variation and Morphological Gradient Priors with Optimization of Structuring Element

    Shoya Oohara, Hirotaka Oka, Mitsuji Muneyasu, Soh Yoshida, Makoto Nakashizuka

    2018 International Symposium on Intelligent Signal Processing and Communication Systems   2018.11

     More details

  • Video Retrieval by Reranking and Relevance Feedback with Tag-Based Similarity

    Takamasa Fujii, Soh Yoshida, Mitsuji Muneyasu

    2018 IEEE 7th Global Conference on Consumer Electronics   2018.9

     More details

  • 異種メタデータを用いたグラフベースソフトクラスタリングに基づくWeb映像リランキング

    吉田 壮, 棟安 実治

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2018.6

     More details

  • タグベクトル表現を用いたWeb映像検索結果のリランキング

    藤井 孝匡, 吉田 壮, 棟安 実治

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2018.6

     More details

  • 複数構造要素とその最適化を用いたモルフォロジカル勾配に基づく画像の正則化

    岡 広高, 大原 翔矢, 棟安 実治, 吉田 壮, 中静 真

    電子情報通信学会技術報告,スマートインフォメディアシステム研究会   2018.6

     More details

  • マルチレイヤグラフを用いた映像特徴の統合に基づく映像検索手法の精度向上

    吉田 壮, 中村 健太郎, 棟安 実治

    電気学会研究会資料   2018.1

     More details

  • 特徴点軌跡とパーティクルフィルタを用いた動作認識手法

    吉岡 真一郎, 棟安 実治, 吉田 壮

    電子情報通信学会技術研究報告   2017.12

     More details

  • 任意に配置されたデータ埋め込み画像の検出”,電子情報通信学会技術研究報告

    天見 元紀, 棟安 実治, 吉田 壮

    電子情報通信学会技術研究報告   2017.12

     More details

  • 構造要素の最適化を考慮したモルフォロジカル勾配に基づく画像の正則化の一手法

    池下 雄大, 棟安 実治, 中静 真, 吉田 壮

    電子情報通信学会技術研究報告   2017.12

     More details

  • 2値ホログラフィ干渉縞データの非可逆圧縮の一手法

    西垣内 崇宏, 棟安 実治, 松島 恭治, 吉田 壮, 田口 亮

    電子情報通信学会技術研究報告   2017.10

     More details

  • User-Oriented Video Search Reranking with Relevance Feedback

    2017 ( 21 )   87 - 92   2017.9

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • ユーザフィードバックを利用したコンテンツの適合性に基づく映像検索

    藤井 孝匡, 吉田 壮, 棟安 実治

    電子情報通信学会ソサイエティ大会   2017.9

     More details

  • 特徴点軌跡とパーティクルフィルタによる動作認識の一手法

    吉岡 真一郎, 棟安 実治, 吉田 壮

    電子情報通信学会ソサイエティ大会   2017.9

     More details

  • A New Method of Lossless Coding for Binary Holographic Interference Fringes

    NISHIGAITO,Takahiro, MUNEYASU,Mitsuji, MATSUSHIMA,Kyoji, YOSHIDA,Soh, TAGUCHI,Akitra

    2017 International Workshop on Smart Info-Media Systems in Asia (SISA 2017)   2017.9

     More details

  • Web Video Search Reranking using a Heterogeneous Graph-Based Soft Clustering Approach

    吉田 壮, 小川 貴弘, 長谷山 美紀, 棟安 実治

    画像の認識・理解シンポジウム (MIRU2017)   2017.8

     More details

  • A Design Technique of Impulse Detector Using Neural Network

    ABE,Seiya, MUNEYASU,Mitsujj, YOSHIDA,Soh

    Proceedings of 2017 Taiwan and Japan Conference on Circuits and Systems   2017.8

     More details

  • Image Regularization with Morphological Gradients Priors Considering Optimization of SE

    IKESHITA,Yudai, MUNEYASU,Mitsujj, NAKASHIZUKA,Makoto, YOSHIDA,Soh

    Proceedings of 2017 Taiwan and Japan Conference on Circuits and Systems   2017.8

     More details

  • Information Retrieval and Estimation of Geometrical State of Information Embedded Marker Using Printed Image

    117 ( 70 )   35 - 40   2017.6

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Image Regularization with Morphological Gradient Priors Using Optimization of Structure Element

    117 ( 70 )   13 - 18   2017.6

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Feedback assisted multi-modality reranking for video search improvement

    116 ( 464 )   91 - 95   2017.2

     More details

    Language:Japanese  

    researchmap

  • Feedback assisted multi-modality reranking for video search improvement

    41 ( 5 )   91 - 95   2017.2

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Lossy Coding of Wave Field Data Using Singular Value Decomposition

    116 ( 344 )   13 - 18   2016.12

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Heterogeneous Graph-Based Topic Learning for Web Video Search Reranking

    YOSHIDA,Soh, OGAWA,Takahiro, HASEYAMA,Miki, MUNEYASU,Mitsujj

    2016 International Workshop on Smart Info-Media Systems in Asia (SISA 2016)   2016.9

     More details

  • Improved Object Detection Method Using PTZ Camera and Kinect

    SHINTANI,Takeshi, MUNEYASU,Mitsuji, YOSHIDA,Soh

    2016 International Workshop on Smart Info-Media Systems in Asia (SISA 2016)   2016.9

     More details

  • Improving Performance of Graph-Based Web Video Search Reranking

    40 ( 6 )   197 - 200   2016.2

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Improving Performance of Graph-Based Web Video Search Reranking

    吉田壮, 小川貴弘, 長谷山美紀

    電子情報通信学会技術研究報告   115 ( 458(ITS2015 56-83) )   2016

  • Web映像検索のためのグラフマイニングを用いたリランキング手法

    吉田壮, 小川貴弘, 長谷山美紀, 棟安実治

    信号処理シンポジウム講演論文集(CD-ROM)   31st   2016

  • Web映像検索を目的としたリランキングの高精度化に関する検討

    吉田壮, 小川貴弘, 長谷山美紀, 棟安実治

    電子情報通信学会大会講演論文集(CD-ROM)   2016   2016

  • A note on Web video retrieval based on network structure analysis using video features and tag information : Retrieval performance improvement using local learning regularization

    114 ( 459 )   77 - 82   2015.2

     More details

    Language:Japanese  

    researchmap

  • A note on Web video retrieval based on network structure analysis using video features and tag information : Retrieval performance improvement using local learning regularization

    39 ( 7 )   77 - 82   2015.2

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • グラフ構造に基づくリランキングを利用したWeb映像検索に関する検討~初期検索結果の誤りを考慮した最適化に基づく高精度化~

    吉田壮, 小川貴弘, 長谷山美紀

    信号処理シンポジウム講演論文集(CD-ROM)   30th   2015

  • A Note on Improving Video Scene Segmentation Performance Using Structure Analysis Based on Hidden Conditional Random Fields

    113 ( 434 )   285 - 290   2014.2

     More details

    Language:Japanese  

    researchmap

  • A Note on Improving Video Scene Segmentation Performance Using Structure Analysis Based on Hidden Conditional Random Fields

    38 ( 7 )   285 - 290   2014.2

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Effectiveness Verification of Audio-based Preprocessing on a Scene Segmentation Method in Music Programs

    YOSHIDA,Soh

    2013 International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC 2013)   2013.7

     More details

  • A Scene Segmentation Method Based on Video Structures in Music Programs : Verifying the Effect of Face Recognition on Scene Segmentation Accuracy

    37 ( 8 )   101 - 106   2013.2

     More details

    Language:Japanese  

    CiNii Books

    researchmap

  • Accurate Scene Segmentation Method Based on Video Structures in Music Programs

    YOSHIDA,Soh

    2013 International Workshop on Advanced Image Technology (IWAIT 2013)   2013.1

     More details

  • Accurate Graph-Based Scene Segmentation for Videos of Music Programs

    YOSHIDA,Soh

    2012 International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC 2012)   2012.7

     More details

  • 物体認識における識別器の高精度化に関する検討 SVMを用いた物体領域の自動選定手法の導入

    吉田壮, 小川貴弘, 長谷山美紀

    電気・情報関係学会北海道支部連合大会講演論文集(CD-ROM)   2012   2012

▼display all

Awards

  • 若手研究優秀賞

    2023.3   電子情報通信学会 スマートインフォメディアシステム研究会  

    曽我 茅冬, 吉田 壮, 棟安 実治

     More details

  • IEEE GCCE 2022 Excellent Poster Award IEEE Consumer Electronics Society

    2022.10   IEEE GCCE 2022   Propagation-Based Fake News Detection Using a Combination of Different Content Features

    Kayato Soga, Soh Yoshida, Mitsuji Muneyasu

     More details

  • 若手研究優秀賞

    2022.3   電子情報通信学会 スマートインフォメディアシステム研究会  

    東本 良太, 吉田 壮, 棟安 実治

     More details

  • 若手研究優秀賞

    2019.3   電子情報通信学会 スマートインフォメディアシステム研究会  

     More details

  • IEEE GCCE 2018 Excellent Poster Award

    2018.10   IEEE Consumer Electronics Society  

    Takamasa Fujii, Soh Yoshida, Mitsuji Muneyasu

     More details

  • 学術奨励賞

    2017.3   電子情報通信学会  

     More details

    Country:Japan

    researchmap

  • 若手奨励賞

    2016.2   電子情報通信学会  

     More details

    Country:Japan

    researchmap

  • 優秀研究発表賞

    2015.12   映像情報メディア学会  

     More details

    Country:Japan

    researchmap

  • 学生員奨励賞

    2014.3   電子情報通信学会  

     More details

    Country:Japan

    researchmap

▼display all

Research Projects

  • Social media recommendation considering the diversity of values

    Grant number:22K18007  2022.4 - 2026.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Early-Career Scientists

      More details

    Grant amount:\4680000 ( Direct Cost: \3600000 、 Indirect Cost:\1080000 )

    researchmap

  • 異なる価値観を融合する検索基盤の創成

    2020.11 - 2022.3

    科学技術振興機構  戦略的創造研究推進事業(ACT-X) 

      More details

    Authorship:Principal investigator 

    researchmap

  • Development of a Data Retrieval Method with Wide Availability Based on Images by Data Embedding and Image Identification

    Grant number:20K04476  2020.4 - 2023.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)  Grant-in-Aid for Scientific Research (C)

      More details

    Grant amount:\4290000 ( Direct Cost: \3300000 、 Indirect Cost:\990000 )

    researchmap

  • 機械工具入札会における画像特徴ベクトル検索システムの構築

    2020 - 2021

    中小企業庁  令和2年度中小企業組合等課題対応支援事業 

    堀川和義

      More details

    Authorship:Collaborating Investigator(s) (not designated on Grant-in-Aid) 

    researchmap

  • 中国出土木簡の書体分析の基礎的研究

    Grant number:19K01044  2019.4 - 2023.3

    日本学術振興会  科学研究費助成事業 基盤研究(C)  基盤研究(C)

    藤田 高夫, 吉田 壮

      More details

    Grant amount:\4420000 ( Direct Cost: \3400000 、 Indirect Cost:\1020000 )

    2021年度は、前年度までの木簡書体分析のために集積した文字データをもとに、木簡からの文字切り出し、木簡文字の字体のくずれ判定、さらに同筆・異筆判定のための基礎的考察を行った。
    本年度の成果として、所有している約4000枚の木簡画像に対し文字領域の位置情報を付与するアノテーションを行うことで、文字検出器の訓練データ不足の問題に対応した。また、昨年度までに開発した文字検出器に、文字の境界部分を検出し分離する後処理を導入した。行方向に接近した文字を正確に検出することで、文字領域誤検出の大幅な削減に成功し、検出精度を向上させた。これにより,文字画像の切り出しの自動化に関して,目処が立ったといえる。
    この成果を踏まえて、木簡文字の字体のくずれを客観的な数値として出力する、くずれ度算出法の開発に着手した。木簡研究者らは従来、隷書の筆跡との差異を、崩れを分類する基準としている。ここでは、これに倣い、隷書文字で構成される文字データから敵対的生成ネットワーク(GAN)を学習し、入力文字とGANが生成する入力文字の再構成文字との間の誤差にもとづいて,崩れ度を算出する手法を開発した。また、木簡字典(佐藤光一著、雄山閣出版、1985)に収録されている文字をもとに、崩れのレベルが異なる5段階の同種文字セットを100組作成し、提案手法の有効性を確認した。
    また、同一文書内での一人の書写者による同一文字の書きぶりにどの程度の偏差を見いだしうるかを探るため、有名な「候粟君所責寇恩事」を例に採りあげ、詳細な検討を加えた。

    researchmap

  • Data retrieval from real environment applicable to augmented reality using projected images by a projector

    Grant number:17K06450  2017.4 - 2020.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)  Grant-in-Aid for Scientific Research (C)

    MUNEYASU Mitsuji

      More details

    Grant amount:\4810000 ( Direct Cost: \3700000 、 Indirect Cost:\1110000 )

    Augmented reality markers are required to be easily placed and to be no discomfort even in the real space. We have succeeded in developing the essential method that is necessary for that purpose, to embed the marker information in the printed image, the projector projected image, and the signage displayed image and to detect the data. In particular, we were able to achieve a high data detection rate for projected images with severe image deterioration. Furthermore, we have developed a method that uses both image identification and data embedding and succeeded in developing an approach with excellent flexibility and robustness by overcoming the drawbacks of both. Also, as an application, we were able to consider a system that uses the logo mark as a base image and gives different information to the same type of marker.

    researchmap

  • Development of methods for video corpus construction and time series annotation for video scene retrieval

    Grant number:17K12687  2017.4 - 2020.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research Grant-in-Aid for Young Scientists (B)  Grant-in-Aid for Young Scientists (B)

    Yoshida Soh

      More details

    Grant amount:\4030000 ( Direct Cost: \3100000 、 Indirect Cost:\930000 )

    We developed methods for supervised video annotation and video scene retrieval based on a corpus constructed from automatically collected learning data from the web. The labels on the corpus are also automatically extracted from the analysis of the web dataset. Furthermore, we developed a reranking method to improve the accuracy of keyword searches to improve the reliability of the training dataset. For developing a video scene retrieval algorithm, we use relevance feedback, which the user can interact with by specifying some example videos from the initial search results. The effectiveness of the proposed methods was confirmed by conducting experiments on the YouTube dataset.

    researchmap

  • 河川管理を支援するCCTV映像自動解析技術に関する研究

    2017 - 2019

    国土交通省 河川砂防技術研究開発公募地域課題分野(河川) 

    小川 貴弘

      More details

    Authorship:Coinvestigator(s) 

    researchmap

▼display all