映像解析AIの最先端技術、XR/メタバース活用の研究開発に取り組んでいます。鉄道、エネルギー、計測分析などの事業ドメインを対象とし、社会イノベーション事業を創生・牽引するとともに、だれもが安心・安全・健康を享受できる社会の実現に貢献します。
コンピュータビジョン、画像認識、行動認識、センサ信号処理、生成AI、マルチモーダル認識、XRコンテンツ生成、UX、感覚拡張、機械学習、知識処理
アプリケーション例:インダストリアルメタバース、映像監視、作業訓練、現場作業支援、計測分析システム、OTナレッジマネジメント、生体認証、RPA、医療応用
2020年から2024年1月現在までの英文での発表は以下の通りです。これ以外のパブリケーションリストは下記のページをご覧下さい。
https://hitachi-speech.github.io/
Onkar Krishna, Hiroki Ohashi, Saptarshi Sinha, MILA: Memory-Based Instance-Level Adaptation for Cross-Domain Object Detection, BMVC 2023.
Michael Lorenz, Rebecca Keilhauer, Takayuki Akiyama, Takehiro Niikura, Didier Stricker, Bertram Taetz, Gabriele Bleser, In Proceedings of the 9th International Conference on Control, Decision and Information Technologies (CoDIT-2023).
Cesar Daniel Rojas Ferrer, Masayoshi Ishibashi, Takayuki Fujiwara, Work-in-Progress—Exploring the Effectiveness of Multi-User Basic Workflow Assessment Training in Web XR, iLRN 2023.
Yoshiki Ito, HOKEM: Human and Object Keypoint-based Extension Module for Human-Object Interaction Detection, ICIP 2023.
Yasuki Kakishita, Hideharu Hattori, Akiko Hisada, Erino Matsumoto, Yusuke Ominami, Jacques Yaacoub Bou Khalil, Didier Raoult, Deep Learning Based Bacteria Classification from SEM Images using Region-wise and Subregion-wise Feature, EMBC 2023.
Saptarshi Sinha, Hiroki Ohashi, Difficulty-Net: Learning to Predict Difficulty for Long-Tailed Recognition, WACV 2023.
Yasuki Kakishita, Hideharu Hattori, Akiko Hisada, Yusuke Ominami, Jean-Pierre Baudoin, Jacques Yaacoub Bou Khalil. Didier Raoult, Deep Learning Based Bacteria Classification from SEM Images Using a Combination of Membrane and Internal Features, CISS 2022.
Yoshiki Ito, Quan Kong, Kenichi Morita, Tomoaki Yoshinaga, Efficient and Accurate Skeleton-Based Two-Person Interaction Recognition Using Inter-and Intra-body Graphs, ICIP 2022.
Masato Tamura, Rahul Vishwakarma, Ravigopal Vennelakanti, Hunting Group Clues with Transformers for Social Group Activity Recognition, ECCV 2022.
Saptarshi Sinha, Hiroki Ohashi, Katsuyuki Nakamura, Class-Difficulty Based Methods for Long-Tailed Visual Recognition, International Journal of Computer Vision (IJCV) 2022.
Ziwei Deng, Quan Kong, Naoto Akira, Tomoaki Yoshinaga, Hierarchical Contrastive Adaptation for Cross-Domain Object Detection, Machine Vision and Applications, 2022.
Sanjay Kumar Dwivedi, Cesar Daniel Rojas Ferrer, Takayuki Fujiwara, Work-in-Progress—Effectiveness of Avatar Guided Instructions and Feedback in the Server Repair and Maintenance Task, iLRN 2022.
Cesar Daniel Rojas Ferrer, Takayuki Fujiwara, Work-in-Progress—Avatar vs Teams: Co-presence and Visual Behavior in a Videoconference Cooperative Task, iLRN 2022.
Yasuki Kakishita, Hideharu Hattori, Akiko Hisada, Yusuke Ominami, Jean-Pierre Baudoin, Jacques Yaacoub Bou Khalil. Didier Raoult, Quantitative Analysis System for Bacterial Cells in SEM Image using Deep Learning, CISS 2021.
Hiroki Ohashi, Hiroto Nagayoshi, Human-error-potential Estimation based on Wearable Biometric Sensors, KDIR 2021.
Katsuyuki Nakamura, Hiroki Ohashi, Mitsuhiro Okada, Sensor-Augmented Egocentric-Video Captioning with Dynamic Modal-Attention, ACM Multimedia, 2021.
C.-H. Huck Yang, Mohit Chhabra, Y.-C. Liu, Quan Kong, Tomoaki Yoshinaga, Tomokazu Murakami, Robust Unsupervised Multi-Object Tracking in Noisy Environments, ICIP 2021.
Yoshiki Ito, Kenichi Morita, Quan Kong, Tomoaki Yoshinaga, Multi-Stream Adaptive Graph Convolutional Network Using Inter- and Intra-Body Graphs for Two-Person Interaction Recognition, IEEE Access, vol. 9, 2021.
Masato Tamura, Hiroki Ohashi, Tomoaki Yoshinaga, QPIC: Query-Based Pairwise Human-Object Interaction Detection with Image-Wide Contextual Information, CVPR 2021.
Naoyuki Terashita, Hiroki Ohashi, Yuichi Nonaka, Takashi Kanemaru, Influence Estimation for Generative Adversarial Networks, ICLR 2021.
Yuki Inoue, Hiroto Nagayoshi, Crack Detection as a Weakly-Supervised Problem: Towards Achieving Less Annotation-Intensive Crack Detectors, ICPR 2021.
Quan Kong, Wenpeng Wei, Ziwei Deng, Tomoaki Yoshinaga, Tomokazu Murakami, Cycle-Contrast for Self-Supervised Video Representation Learning, NeurIPS 2020.
Saptarshi Sinha, Hiroki Ohashi, Katsuyuki Nakamura, Class-Wise Difficulty-Balanced Loss for Solving Class-Imbalance, ACCV 2020.
Masato Tamura, Tomoaki Yoshinaga, BcaR: Beginner Classifier as Regularization Towards Generalizable Re-ID, BMVC 2020.
Koichiro Ito, Quan Kong, Shota Horiguchi, Takashi Sumiyoshi, Kenji Nagamatsu, Anticipating the Start of User Interaction for Service Robot in the Wild, ICRA 2020.
田村雅人、中村克行、「Pythonで学ぶ画像認識 機械学習実践シリーズ」、インプレス、2023年3月
渡邉 裕樹、廣池 敦、吉永 智明、福田 安宏、「安心・安全な社会構築を支える映像解析AI技術」、画像ラボ 2023年1月号、日本工業出版(株)
日立、現場データの収集技術や生成AIを活用した「現場拡張メタバース」を開発:研究開発:日立 (hitachi.co.jp)
AI映像解析技術で監視・警備業務を高度化するソリューション「Hitachi Multifeature Video Search」に不審人物や荷物の置き引きを早期発見するための新機能を拡充:研究開発:日立 (hitachi.co.jp)
南紀白浜空港で、保安検査を支援する人工知能(AI)技術の実用化に向けた実証実験を拡大:研究開発:日立 (hitachi.co.jp)
災害状況をAIで把握する映像解析の基礎技術を開発:研究開発:日立 (hitachi.co.jp)
作業者の身体負荷をスーツ型のウェアラブルデバイスにより定量評価し、作業動作の改善点を提示するAIを開発:研究開発:日立 (hitachi.co.jp)
手のひらと顔で、専用装置を使わず安全・安心な手ぶら決済を実現:研究開発:日立 (hitachi.co.jp)