A Deep Learning-Based Approach to Facilitate Cross-cultural Kansei Design

Xiaofei Zhou, Pei Luen Patrick Rau, Michiko Ohkura, Tipporn Laohakangvalvit, Bingcheng Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

With the development of economic globalization, more and more product designers are faced with the need of designing for customers from different countries. However, it is a challenge for designers to efficiently develop the desired mental images of certain products for target users with different cultural backgrounds. Our research proposed a deep learning-based system to facilitate designers to gain better awareness of the cross-culture differences between different target customers. We trained a kawaii classification neural network model with the data of 1414 cosmetic packaging images annotated by 12 Japanese females separately. As a follow-up investigation, we conducted neuron analysis to compare the features of kawaii packages perceived by Japanese participants with the results from a prior study conducted with Chinese participants. The result shows that Japanese females tended to see more girlish and exquisite design features as kawaii while Chinese females perceived more childish and round elements as kawaii. A reverse experiment further verified the effectiveness of adding these different design features to enhance Chinese or Japanese females’ perception of kawaii. We also noticed that it’s hard to obtain the cross-cultural differences in customers’ perception by extracting image parameters with a set of predefined visual features as such perception differences could be subconscious. Our deep learning-based Kansei design facilitation provides a feasible solution to customized design for target customers with different cultural backgrounds.

Original languageEnglish
Title of host publicationCross-Cultural Design. Interaction Design Across Cultures - 14th International Conference, CCD 2022, Held as Part of the 24th HCI International Conference, HCII 2022, Proceedings
EditorsPei-Luen Patrick Rau
PublisherSpringer Science and Business Media Deutschland GmbH
Pages145-160
Number of pages16
ISBN (Print)9783031060373
DOIs
Publication statusPublished - 2022
Event14th International Conference on Cross-Cultural Design, CCD 2022 Held as Part of the 24th HCI International Conference, HCII 2022 - Virtual, Online
Duration: 2022 Jun 262022 Jul 1

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13311 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th International Conference on Cross-Cultural Design, CCD 2022 Held as Part of the 24th HCI International Conference, HCII 2022
CityVirtual, Online
Period22/6/2622/7/1

Keywords

  • Cross-cultural analysis
  • Kansei Engineering
  • Kawaii perception

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'A Deep Learning-Based Approach to Facilitate Cross-cultural Kansei Design'. Together they form a unique fingerprint.

Cite this