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

Method for aesthetic design based on customer's kansei evaluation using deep learning

Research Project

Project/Area Number 20K12540
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 90010:Design-related
Research InstitutionToyota Technological Institute

Principal Investigator

Kobayashi Masakazu  豊田工業大学, 工学部, 准教授 (40409652)

Project Period (FY) 2020-04-01 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2022: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2021: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2020: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Keywords感性工学 / 意匠設計 / 敵対的生成ネットワーク / 生成AI / クラス活性化マッピング / 最適設計 / 深層学習 / ラフ集合 / ファインチューニング / アクティブラーニング / CNN(畳み込みニューラルネットワーク) / GAN(敵対的生成ネットワーク) / CAM(クラス活性化マッピング) / セマンティックセグメンテーション / 畳み込みニューラルネットワーク
Outline of Research at the Start

感性工学の研究分野では,顧客アンケートに基づいて製品意匠と顧客感性の対応関係を分析し,意匠設計に利用することで,設計者によらずに,顧客の感性的要求を満たす製品意匠を設計する方法を研究してきた.本研究ではこれを実現するために,CNN(畳み込みニューラルネットワーク)を用いて製品意匠と顧客感性の対応関係を分析し,分析結果を基にGAN(敵対的生成ネットワーク)を用いて新しい製品意匠を生成するという創成型意匠設計法の検討を行う.

Outline of Final Research Achievements

To achieve aesthetic design that does not rely on designers but is based on customer Kansei evaluations, this research developed new aesthetic design methods based on deep learning techniques. As a result, many design methods have been developed, such as methods for generating product images that reflect customer preferences using Generative Adversarial Network (GAN) and image generation AI, and optimal design methods for generating the products that best suit customers by analyzing the relationships between product aesthetic design and customer preferences using Class Activation Mapping (CAM) and generative AI.

Academic Significance and Societal Importance of the Research Achievements

デザイナーの知識・経験に基づくのではなく,顧客の感性評価に基づく意匠設計を可能にする複数の設計法を提案することができた.これらの方法は,それぞれ特徴があり,状況に応じて使い分けることができる.深層学習を用いることで,製品意匠の人手による分析(前処理)を必要とせず,製品画像を画像のまま利用できる点も提案手法の利点である.また,この1,2年で急激に発展,普及した生成AIや画像生成AIを意匠設計に利用できることを示した点も本研究の意義の一つである.

Report

(5 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • 2021 Research-status Report
  • 2020 Research-status Report
  • Research Products

    (7 results)

All 2024 2023 2022 2021

All Journal Article (3 results) (of which Peer Reviewed: 3 results,  Open Access: 2 results) Presentation (4 results) (of which Int'l Joint Research: 2 results)

  • [Journal Article] Design Generation Using Stable Diffusion and Questionnaire Survey2024

    • Author(s)
      Kobayashi Masakazu、Kume Katsuyoshi
    • Journal Title

      Computer-Aided Design and Applications

      Volume: 21 Pages: 859-868

    • DOI

      10.14733/cadaps.2024.859-868

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Open Access
  • [Journal Article] Generation of Product Design Using Gan Based on Customer's Kansei Evaluation2022

    • Author(s)
      Masakazu Kobayshi, Pongsasit Thongpramoon
    • Journal Title

      Proceedings of 9th International Conference on Kansei Engineering and Emotion Research 2022

      Volume: -

    • Related Report
      2022 Research-status Report
    • Peer Reviewed
  • [Journal Article] Aesthetic Design Based on the Analysis of Questionnaire Results Using Deep Learning Techniques2022

    • Author(s)
      [1]Kobayashi, M., Fujita, S. and Wada, T.
    • Journal Title

      Computer-Aided Design & Applications

      Volume: 19 Issue: 3 Pages: 602-611

    • DOI

      10.14733/cadaps.2022.602-611

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Open Access
  • [Presentation] 画像生成 AI を用いた意匠設計法の検討2023

    • Author(s)
      小林正和,久米克佳
    • Organizer
      第33回設計工学・システム部門講演会
    • Related Report
      2023 Annual Research Report
  • [Presentation] 顧客の感性評価に基づく製品意匠設計2022

    • Author(s)
      小林正和
    • Organizer
      日本応用数理学会 第19回 研究部会連合発表会
    • Related Report
      2022 Research-status Report
  • [Presentation] Optimal design of product aesthetics based on rough set theory and deep learning techniques2022

    • Author(s)
      Masakazu Kobayashi, Yuki Orii
    • Organizer
      Asian Congress of Structural and Multidisciplinary Optimization 2022
    • Related Report
      2022 Research-status Report
    • Int'l Joint Research
  • [Presentation] Aesthetic Design Based on the Analysis of Questionnaire Results Using Deep Learning Techniques2021

    • Author(s)
      Masakazu Kobayashi
    • Organizer
      CAD'21
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research

URL: 

Published: 2020-04-28   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi