• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Developing a Design Methodology of Deep Neural Networks to Accelerate Paradigm Shit Brought by Deep Learning

Research Project

Project/Area Number 19H01110
Research Category

Grant-in-Aid for Scientific Research (A)

Allocation TypeSingle-year Grants
Section一般
Review Section Medium-sized Section 61:Human informatics and related fields
Research InstitutionTohoku University

Principal Investigator

Okatani Takayuki  東北大学, 情報科学研究科, 教授 (00312637)

Co-Investigator(Kenkyū-buntansha) 菅沼 雅徳  東北大学, 情報科学研究科, 助教 (00815813)
Project Period (FY) 2019-04-01 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥42,640,000 (Direct Cost: ¥32,800,000、Indirect Cost: ¥9,840,000)
Fiscal Year 2022: ¥9,620,000 (Direct Cost: ¥7,400,000、Indirect Cost: ¥2,220,000)
Fiscal Year 2021: ¥11,570,000 (Direct Cost: ¥8,900,000、Indirect Cost: ¥2,670,000)
Fiscal Year 2020: ¥11,960,000 (Direct Cost: ¥9,200,000、Indirect Cost: ¥2,760,000)
Fiscal Year 2019: ¥9,490,000 (Direct Cost: ¥7,300,000、Indirect Cost: ¥2,190,000)
Keywords深層学習 / コンピュータビジョン / 人工知能 / ニューラルネットワーク / cv / 画像認識 / 画像処理 / ディープラーニング / 構造自動設計
Outline of Research at the Start

深層学習は,今やAIの範囲を大きく超えて多様な分野で活用され始め,工学の問題解決の 方法論を一変させるパラダイムシフトをもたらしつつある.すなわち,問題解決は,学習データを用意することと,問題に適したネットワーク構造をうまくデザインすることで達成される.残された課題は,ネットワーク構造のデザインに系統的な方法論がなく,試行錯誤に頼らざるを得ないことである.計画では,従来の考え方-ネットワーク構造を手でデザインし,結合の重みを学習で自動決定する-を見直し,デザインと学習の関係を再定義する.様々な取り組みを通じて,深層ネットワーク設計理論の構築を目指す.

Outline of Final Research Achievements

Deep learning requires developers to design a network architecture that is appropriate for each individual task that they want to solve, but no methodology or guidelines have been established for designing good architectures. We aimed to solve this issue by designing architectures that achieve high performance in various tasks. We have successfully developed networks that achieve the highest accuracy (at the time of publication) for various tasks, including image restoration, image understanding, 3D geometry estimation, uncertainty estimation, and self-supervised feature learning. While the results have significant impact for each of the targeted tasks, their integration provide a foundation toward establishing the methodology of neural architectural design.

Academic Significance and Societal Importance of the Research Achievements

深層学習は,近年発展著しい人工知能の中核技術であるとともに,その他の工学やサイエンスにも大きな影響を与えつつある.その一方で,深層ニューラルネットワークの構造設計に確たる方法論がないという課題があった.本研究は,様々な応用ごとに優れた性能を発揮するネットワーク構造の研究を通じて,それぞれの応用問題の解決に貢献するとともに,ネットワーク構造に関する新たな知見を多く生み出した.これらの成果は,構造設計の方法論の基盤を与えている.

Report

(6 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Annual Research Report
  • 2020 Annual Research Report
  • 2019 Comments on the Screening Results   Annual Research Report
  • Research Products

    (32 results)

All 2022 2021 2020 2019 Other

All Int'l Joint Research (1 results) Journal Article (23 results) (of which Int'l Joint Research: 5 results,  Peer Reviewed: 22 results,  Open Access: 19 results) Presentation (7 results) (of which Invited: 6 results) Book (1 results)

  • [Int'l Joint Research] Hong Kong Polytechnic University(中国)

    • Related Report
      2021 Annual Research Report
  • [Journal Article] Symmetry-aware Neural Architecture for Embodied Visual Exploration2022

    • Author(s)
      Liu Shuang、Okatani Takayuki
    • Journal Title

      Proceedings of Computer Vision and Pattern Recognition

      Volume: - Pages: 17221-17230

    • DOI

      10.1109/cvpr52688.2022.01673

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Bridging the Gap from Asymmetry Tricks to Decorrelation Principles in Non-contrastive Self- supervised Learning2022

    • Author(s)
      Kang-Jun Liu, Masanori Suganuma, Takayuki Okatani
    • Journal Title

      Advances in Neural Information Processing Systems 35 (NeurIPS 2022)

      Volume: - Pages: 19824-19835

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Real-time vehicle identification using two-step LSTM method for acceleration-based bridge weigh-in-motion system2022

    • Author(s)
      Zhu Yanjie、Sekiya Hidehiko、Okatani Takayuki、Yoshida Ikumasa、Hirano Shuichi
    • Journal Title

      Journal of Civil Structural Health Monitoring

      Volume: 12 Issue: 3 Pages: 689-703

    • DOI

      10.1007/s13349-022-00576-2

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] A lightweight deep learning model for automatic segmentation and analysis of ophthalmic images2022

    • Author(s)
      Sharma Parmanand、Ninomiya Takahiro、Omodaka Kazuko、Takahashi Naoki、Miya Takehiro、Himori Noriko、Okatani Takayuki、Nakazawa Toru
    • Journal Title

      Scientific Reports

      Volume: 12 Issue: 1 Pages: 1-18

    • DOI

      10.1038/s41598-022-12486-w

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Deep learning increases the availability of organism photographs taken by citizens in citizen science programs2022

    • Author(s)
      Suzuki-Ohno Yukari、Westfechtel Thomas、Yokoyama Jun、Ohno Kazunori、Nakashizuka Tohru、Kawata Masakado、Okatani Takayuki
    • Journal Title

      Scientific Reports

      Volume: 12 Issue: 1 Pages: 1-10

    • DOI

      10.1038/s41598-022-05163-5

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Matching in the Dark: A Dataset for Matching Image Pairs of Low-light Scenes2021

    • Author(s)
      Wenzheng Song, Masanori Suganuma, Xing Liu, Noriyuki Shimobayashi, Daisuke Maruta, Takayuki Okatani
    • Journal Title

      Proceedings of International Conference on Computer Visionツ?2021

      Volume: - Pages: 6009-6018

    • DOI

      10.1109/iccv48922.2021.00597

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Progressive and Selective Fusion Network for High Dynamic Range Imaging2021

    • Author(s)
      Qian Ye, Jun Xiao, Kin-Man Lam, Takayuki Okatani
    • Journal Title

      Proceedings of ACM Multimedia 2021

      Volume: - Pages: 5290-5297

    • DOI

      10.1145/3474085.3475651

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Learning to Bundle-adjust: A Graph Network Approach to Faster Optimization of Bundle Adjustment for Vehicular SLAM2021

    • Author(s)
      Tetsuya Tanaka, Yukihiro Sasagawa, Takayuki Okatani
    • Journal Title

      Proceedings of International Conference on Computer Vision 2021

      Volume: - Pages: 6320-6329

    • DOI

      10.1109/iccv48922.2021.00619

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Look Wide and Interpret Twice: Improving Performance on Interactive Instruction-following Tasks2021

    • Author(s)
      Van-Quang Nguyen, Masanori Suganuma, Takayuki Okatani
    • Journal Title

      Proceedings of 30th International Joint Conference on Artificial Intelligence (IJCAI-21)

      Volume: - Pages: 923-930

    • DOI

      10.24963/ijcai.2021/128

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] Hyperparameter-Free Out-of-Distribution Detection Using Cosine Similarity2020

    • Author(s)
      Engkarat Techapanurak, Masanori Suganuma, Takayuki Okatani:
    • Journal Title

      Proceedings of Asian Conference on Computer Vision

      Volume: - Pages: 53-69

    • DOI

      10.1007/978-3-030-69538-5_4

    • ISBN
      9783030695378, 9783030695385
    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Removal of Image Obstacles for Vehicle-mounted Surrounding Monitoring Cameras by Real-time Video Inpainting2020

    • Author(s)
      Yoshihiro Hirohashi, Kenichi Narioka, Masanori Suganuma, Xing Liu, Yukimasa Tamatsu, Takayuki Okatani
    • Journal Title

      Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops

      Volume: - Pages: 857-866

    • DOI

      10.1109/cvprw50498.2020.00115

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Efficient Attention Mechanism for Visual Dialog that Can Handle All the Interactions Between Multiple Inputs2020

    • Author(s)
      Van-Quang Nguyen, Masanori Suganuma, Takayuki Okatani
    • Journal Title

      Proceedings of European Conference on Computer Vision

      Volume: - Pages: 223-240

    • DOI

      10.1007/978-3-030-58586-0_14

    • ISBN
      9783030585853, 9783030585860
    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Extending information maximization from a rate-distortion perspective2020

    • Author(s)
      Yan Zhang, Junjie Hu, Takayuki Okatani
    • Journal Title

      Neurocomputing

      Volume: 399 Pages: 285-295

    • DOI

      10.1016/j.neucom.2020.02.061

    • Related Report
      2020 Annual Research Report 2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Analysis and a Solution of Momentarily Missed Detection for Anchor-based Object Detectors2020

    • Author(s)
      Yusuke Hosoya, Masanori Suganuma, Takayuki Okatani
    • Journal Title

      Proceedings of IEEE Winter Conference on Applications of Computer Vision

      Volume: - Pages: 1399-1407

    • DOI

      10.1109/wacv45572.2020.9093553

    • Related Report
      2020 Annual Research Report 2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Toward Explainable Fashion Recommendation2020

    • Author(s)
      Pongsate Tangseng, Takayuki Okatani
    • Journal Title

      Proceedings of IEEE Winter Conference on Applications of Computer Vision

      Volume: - Pages: 2142-2151

    • DOI

      10.1109/wacv45572.2020.9093367

    • Related Report
      2020 Annual Research Report 2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Joint Learning of Multiple Image Restoration Tasks2020

    • Author(s)
      Xing Liu, Masanori Suganuma, Takayuki Okatani
    • Journal Title

      arXiv.org

      Volume: - Pages: 1-10

    • Related Report
      2020 Annual Research Report
    • Open Access
  • [Journal Article] Deep Learning for Natural Image Reconstruction from Electrocorticography Signals2019

    • Author(s)
      Date Hiroto、Kawasaki Keisuke、Hasegawa Isao、Okatani Takayuki
    • Journal Title

      Proceedings of BIBM

      Volume: - Pages: 2331-2336

    • DOI

      10.1109/bibm47256.2019.8983029

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Dual Residual Networks Leveraging the Potential of Paired Operations for Image Restoration2019

    • Author(s)
      Liu Xing、Suganuma Masanori、Sun Zhun、Okatani Takayuki
    • Journal Title

      Proceedings of IEEE Computer Vision and Pattern Recognition

      Volume: - Pages: 7007-7016

    • DOI

      10.1109/cvpr.2019.00717

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Attention-Based Adaptive Selection of Operations for Image Restoration in the Presence of Unknown Combined Distortions2019

    • Author(s)
      Suganuma Masanori、Liu Xing、Okatani Takayuki
    • Journal Title

      Proceedings of IEEE Computer Vision and Pattern Recognition

      Volume: - Pages: 9039-9048

    • DOI

      10.1109/cvpr.2019.00925

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Multi-Task Learning of Hierarchical Vision-Language Representation2019

    • Author(s)
      Nguyen Duy-Kien、Okatani Takayuki
    • Journal Title

      Proceedings of IEEE Computer Vision and Pattern Recognition

      Volume: - Pages: 10492-10501

    • DOI

      10.1109/cvpr.2019.01074

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Improving Head Pose Estimation with a Combined Loss and Bounding Box Margin Adjustment2019

    • Author(s)
      Shao Mingzhen、Sun Zhun、Ozay Mete、Okatani Takayuki
    • Journal Title

      Proceedings of FG

      Volume: - Pages: 1-5

    • DOI

      10.1109/fg.2019.8756605

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Visualization of Convolutional Neural Networks for Monocular Depth Estimation2019

    • Author(s)
      Hu Junjie、Zhang Yan、Okatani Takayuki
    • Journal Title

      Proceedings of International Conference on Computer Vision

      Volume: - Pages: 3868-3877

    • DOI

      10.1109/iccv.2019.00397

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] A Generative Model of Underwater Images for Active Landmark Detection and Docking2019

    • Author(s)
      Liu Shuang、Ozay Mete、Xu Hongli、Lin Yang、Okatani Takayuki
    • Journal Title

      Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems

      Volume: - Pages: 8034-8039

    • DOI

      10.1109/iros40897.2019.8968146

    • Related Report
      2019 Annual Research Report
    • Peer Reviewed
  • [Presentation] 画像を理解し,その内容を人と共有できるAIの実現へ向けて 3.学会等名 土木学会講演会(招待講演) 4.発表年 2022年 1.発表者名 岡谷貴之 提出2022

    • Author(s)
      岡谷貴之
    • Organizer
      土木学会講演会
    • Related Report
      2022 Annual Research Report
    • Invited
  • [Presentation] 画像を扱うAI(≒深層学習)に関する研究2022

    • Author(s)
      岡谷貴之
    • Organizer
      仙台X-Tech
    • Related Report
      2022 Annual Research Report
    • Invited
  • [Presentation] 深層学習の現在と近未来:深奥質感からAIの今後を考える2021

    • Author(s)
      岡谷貴之
    • Organizer
      新道路成果報告会AI活用
    • Related Report
      2021 Annual Research Report
    • Invited
  • [Presentation] 深層学習の現在:問題解決の方法論として2021

    • Author(s)
      岡谷貴之
    • Organizer
      日本天文学会
    • Related Report
      2021 Annual Research Report
    • Invited
  • [Presentation] ディープラーニングの課題: 現場からフロンティアまで2021

    • Author(s)
      岡谷貴之
    • Organizer
      日本機会学会
    • Related Report
      2021 Annual Research Report
    • Invited
  • [Presentation] 深層学習(≒AI)の現在と近い将来2021

    • Author(s)
      岡谷貴之
    • Organizer
      精密工学会
    • Related Report
      2021 Annual Research Report
    • Invited
  • [Presentation] 言語による画像理解とそれに基づく行動の生成まで2021

    • Author(s)
      岡谷貴之
    • Organizer
      CRESTセミナー
    • Related Report
      2021 Annual Research Report
  • [Book] 深層学習 改訂第2版2022

    • Author(s)
      岡谷 貴之
    • Total Pages
      384
    • Publisher
      講談社
    • ISBN
      4065133327
    • Related Report
      2022 Annual Research Report 2021 Annual Research Report

URL: 

Published: 2019-04-18   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi