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

Development of Technology for Realizing High-Compression of Video by Extracting Important Regions

Research Project

Project/Area Number 20H04201
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Review Section Basic Section 61010:Perceptual information processing-related
Research InstitutionTohoku University

Principal Investigator

Omachi Shinichiro  東北大学, 工学研究科, 教授 (30250856)

Co-Investigator(Kenkyū-buntansha) 宮崎 智  東北大学, 工学研究科, 助教 (10755101)
大町 方子  仙台高等専門学校, 総合工学科, 教授 (90316448)
菅谷 至寛  東北大学, 工学研究科, 准教授 (80323062)
Project Period (FY) 2020-04-01 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥17,290,000 (Direct Cost: ¥13,300,000、Indirect Cost: ¥3,990,000)
Fiscal Year 2022: ¥7,150,000 (Direct Cost: ¥5,500,000、Indirect Cost: ¥1,650,000)
Fiscal Year 2021: ¥7,020,000 (Direct Cost: ¥5,400,000、Indirect Cost: ¥1,620,000)
Fiscal Year 2020: ¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
Keywords映像符号化
Outline of Research at the Start

日々爆発的に生成され続けるデータの量は保存できるストレージの量をはるかに超えており、膨大なデータの中から真に必要なデータのみを残していく技術を開発することは喫緊の課題である。本研究課題では、映像データを対象とし、映像から重要領域を検出し、映像としての価値や必要な情報を損なわずに高圧縮を実現する技術を開発する。

Outline of Final Research Achievements

The purpose of this research was to develop a technology for detecting important regions in video and achieving high compression without losing the value of the video or necessary information. Unlike the conventional trend of video coding that preserves the quality of the entire image, the main characteristic of this technology is that it aims for high compression without losing the value and necessary information as a video. As a result, we developed a technology for detecting important regions in images, a technology for evaluating image visibility, and a high-compression image coding technology using machine learning.

Academic Significance and Societal Importance of the Research Achievements

人が生成するデータの量は爆発的に増加している。その中でも映像データはインターネットトラフィックの8割を占めており、効率的な保存や伝送が求められている。本研究は画像中の各領域の重要度の推定法や視認性を評価する指標を開発し、映像中から真に必要な情報を残しつつ高圧縮するための手法を開発したものである。映像通信の効率を向上させることが可能となり、様々な場面において映像データを有効活用するための基盤技術となり得る。

Report

(4 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Annual Research Report
  • 2020 Annual Research Report
  • Research Products

    (31 results)

All 2023 2022 2021 2020

All Journal Article (11 results) (of which Peer Reviewed: 10 results,  Open Access: 6 results) Presentation (19 results) (of which Int'l Joint Research: 9 results,  Invited: 2 results) Book (1 results)

  • [Journal Article] 情景画像中の重要テキストの検出と伝送2023

    • Author(s)
      大町真一郎, 宮崎 智
    • Journal Title

      画像ラボ

      Volume: 34 Pages: 17-22

    • Related Report
      2022 Annual Research Report
  • [Journal Article] Important Region Estimation Using Image Captioning2022

    • Author(s)
      Taku Suzuki, Daisuke Sato, Yoshihiro Sugaya, Tomo Miyazaki, and Shinichiro Omachi
    • Journal Title

      IEEE Access

      Volume: 10 Pages: 105546-105555

    • DOI

      10.1109/access.2022.3211260

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Correlation-Based Data Augmentation for Machine Learning and its Application to Road Environment Recognition2022

    • Author(s)
      Shinichiro Omachi and Masako Omachi
    • Journal Title

      IEEE Transactions on Vehicular Technology

      Volume: -

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Importance Estimation for Scene Texts Using Visual Features2022

    • Author(s)
      Kota Oodaira, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Journal Title

      Interdisciplinary Information Sciences

      Volume: -

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Stroke-Based Scene Text Erasing Using Synthetic Data for Training2021

    • Author(s)
      Zhengmi Tang, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Journal Title

      IEEE Transactions on Image Processing

      Volume: 30 Pages: 9306-9320

    • DOI

      10.1109/tip.2021.3125260

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] A Framework for Estimating Gaze Point Information for Location-Based Services2021

    • Author(s)
      Junpei Masuho, Tomo Miyazaki, Yoshihiro Sugaya, Masako Omachi, Shinichiro Omachi
    • Journal Title

      IEEE Transactions on Vehicular Technology

      Volume: 70 Issue: 9 Pages: 8468-8477

    • DOI

      10.1109/tvt.2021.3101932

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Multiple Visual-Semantic Embedding for Video Retrieval from Query Sentence2021

    • Author(s)
      Nguyen Huy Manh、Miyazaki Tomo、Sugaya Yoshihiro、Omachi Shinichiro
    • Journal Title

      Applied Sciences

      Volume: 11 Issue: 7 Pages: 3214-3214

    • DOI

      10.3390/app11073214

    • Related Report
      2021 Annual Research Report 2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Self Texture Transfer Networks for Low Bitrate Image Compression2021

    • Author(s)
      Shoma Iwai, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Journal Title

      Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops

      Volume: - Pages: 1901-1905

    • DOI

      10.1109/cvprw53098.2021.00214

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Mackerel Fat Content Estimation Using RGB and Depth Images2021

    • Author(s)
      Sano Shuya、Miyazaki Tomo、Sugaya Yoshihiro、Sekiguchi Naohiro、Omachi Shinichiro
    • Journal Title

      IEEE Access

      Volume: 9 Pages: 164060-164069

    • DOI

      10.1109/access.2021.3134260

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Text Detection Using Multi-Stage Region Proposal Network Sensitive to Text Scale2021

    • Author(s)
      Yoshito Nagaoka, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Journal Title

      Sensors

      Volume: 21 Issue: 4 Pages: 1232-1232

    • DOI

      10.3390/s21041232

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Fidelity-Controllable Extreme Image Compression with Generative Adversarial Networks2021

    • Author(s)
      Shoma Iwai, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Journal Title

      Proceedings of the 25th International Conference on Pattern Recognition

      Volume: - Pages: 8235-8242

    • Related Report
      2020 Annual Research Report
    • Peer Reviewed
  • [Presentation] Masked Image Modelingを利用した情景画像中のテキスト認識2023

    • Author(s)
      三ツ井悠翔, 宮崎智, 大町真一郎
    • Organizer
      言語処理学会第29回年次大会
    • Related Report
      2022 Annual Research Report
  • [Presentation] Detecting Important Information from Scene Images via Texts2023

    • Author(s)
      Shinichiro Omachi
    • Organizer
      The 6th Tohoku Uni-NTU Symposium on Interdisciplinary AI and Human Studies
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Rethinking Degradation: Radiograph Super-Resolution via AID-SRGAN2022

    • Author(s)
      Yongsong Huang, Qingzhong Wang, Shinichiro Omachi
    • Organizer
      The 13th International Workshop on Machine Learning in Medical Imaging
    • Related Report
      2022 Annual Research Report
    • Int'l Joint Research
  • [Presentation] GANを使った画像符号化における識別器に関する検討2022

    • Author(s)
      岩井翔真, 宮崎智, 大町真一郎
    • Organizer
      画像の認識・理解シンポジウム2022
    • Related Report
      2022 Annual Research Report
  • [Presentation] 物体のテクスチャに注目したGANによる熱赤外線画像着色手法の検討2022

    • Author(s)
      宇川慧, 宮崎智, 大町真一郎
    • Organizer
      画像の認識・理解シンポジウム2022
    • Related Report
      2022 Annual Research Report
  • [Presentation] 編集距離に基づいた距離学習による情景画像文字認識結果の修正に関する検討2022

    • Author(s)
      梶原颯希, 宮崎智, 大町真一郎
    • Organizer
      画像の認識・理解シンポジウム2022
    • Related Report
      2022 Annual Research Report
  • [Presentation] U-Netを用いた電子部品画像からの配線検出2022

    • Author(s)
      谷本衡哉, 宮崎智, 大町真一郎
    • Organizer
      画像の認識・理解シンポジウム2022
    • Related Report
      2022 Annual Research Report
  • [Presentation] Image and Video Compression Preserving Important Information for Humans2021

    • Author(s)
      Shinichiro Omachi
    • Organizer
      2021 6th International Conference on Signal and Image Processing (ICSIP 2021)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Self Texture Transfer Networks for Low Bitrate Image Compression2021

    • Author(s)
      Shoma Iwai, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Organizer
      The 4th Workshop and Challenge on Learned Image Compression
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Semantic Guided Extreme Image Compression with Generative Adversarial Networks2021

    • Author(s)
      Shoma Iwai, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Organizer
      電気関係学会東北支部連合大会
    • Related Report
      2021 Annual Research Report
  • [Presentation] Erasing Scene Text Based on Stroke Prediction2021

    • Author(s)
      Zhengmi Tang, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Organizer
      電気関係学会東北支部連合大会
    • Related Report
      2021 Annual Research Report
  • [Presentation] 文字の可読性を考慮した画像の高効率符号化手法の実験的検証2021

    • Author(s)
      内ヶ崎翔平, 菅谷至寛, 宮崎智, 大町真一郎
    • Organizer
      画像の認識・理解シンポジウム
    • Related Report
      2021 Annual Research Report
  • [Presentation] 参照型超解像による特徴転移を利用した高品質な画像符号化2021

    • Author(s)
      岩井翔真, 宮崎智, 菅谷至寛, 大町真一郎
    • Organizer
      画像の認識・理解シンポジウム
    • Related Report
      2021 Annual Research Report
  • [Presentation] Extraction of important text information from images2021

    • Author(s)
      Shinichiro Omachi
    • Organizer
      Symposium of Yotta Informatics Research Platform for Yotta-Scale Data Science 2021
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Fidelity-Controllable Extreme Image Compression with Generative Adversarial Networks2021

    • Author(s)
      Shoma Iwai, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Organizer
      25th International Conference on Pattern Recognition
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Study of Detecting Important Regions in Natural Images with Image Captioning2020

    • Author(s)
      Taku Suzuki, Daisuke Sato, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Organizer
      The 17th International Workshop on Emerging ICT
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Survey of Automatic Video Colorization by DeepNeural Network2020

    • Author(s)
      Takuto Toida, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Organizer
      The 17th International Workshop on Emerging ICT
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Two-Stage Training for High-Fidelity Image Compression with Generative Adversarial Networks2020

    • Author(s)
      Shoma Iwai, Tomo Miyazaki, Yoshihiro Sugaya, Shinichiro Omachi
    • Organizer
      The 17th International Workshop on Emerging ICT
    • Related Report
      2020 Annual Research Report
    • Int'l Joint Research
  • [Presentation] キャプショニングを用いた画像中の重要領域判別手法の検討2020

    • Author(s)
      鈴木 拓, 佐藤大亮, 宮崎 智, 菅谷至寛, 大町真一郎
    • Organizer
      第23回 画像の認識・理解シンポジウム
    • Related Report
      2020 Annual Research Report
  • [Book] 機械学習の可能性2022

    • Author(s)
      計測自動制御学会、浮田 浩行、濱上 知樹、藤吉 弘亘、大町 真一郎、戸田 智基、岩崎 敦、小林 泰介、鈴木 亮太、木村 雄喜、橋本 大樹、玉垣 勇樹、水谷 麻紀子、永田 毅、木村 光成、李 晃伸、川嶋 宏彰
    • Total Pages
      240
    • Publisher
      コロナ社
    • ISBN
      9784339033854
    • Related Report
      2022 Annual Research Report

URL: 

Published: 2020-04-28   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi