DOI QR코드

DOI QR Code

패션 AI의 학습 데이터 표준화를 위한 패션 아이템 이미지의 색채와 소재 속성 분류 체계

Color & Texture Attribute Classification System of Fashion Item Image for Standardizing Learning Data in Fashion AI

  • Park, Nanghee (Dept. of Clothing & Textiles, Chungnam National University) ;
  • Choi, Yoonmi (Dept. of Clothing & Textiles, Chungnam National University)
  • 투고 : 2020.02.26
  • 심사 : 2020.03.31
  • 발행 : 2020.04.30

초록

Accurate and versatile image data-sets are essential for fashion AI research and AI-based fashion businesses based on a systematic attribute classification system. This study constructs a color and texture attribute hierarchical classification system by collecting fashion item images and analyzing the metadata of fashion items described by consumers. Essential dimensions to explain color and texture attributes were extracted; in addition, attribute values for each dimension were constructed based on metadata and previous studies. This hierarchical classification system satisfies consistency, exclusiveness, inclusiveness, and flexibility. The image tagging to confirm the usefulness of the proposed classification system indicated that the contents of attributes of the same image differ depending on the annotator that require a clear standard for distinguishing differences between the properties. This classification system will improve the reliability of the training data for machine learning, by providing standardized criteria for tasks such as tagging and annotating of fashion items.

키워드

참고문헌

  1. An, H., Kwon, S., & Park, M. (2019). A case study on the recommendation services for customized fashion styles based on artificial intelligence. Journal of the Korean Society of Clothing and Textiles, 43(3), 349-360. doi:10.5850/JKSCT.2019.43.3.349
  2. Arnheim, R. (2004). Art and visual perception: A psychology of the creative eye (The new version). Berkeley, Los Angeles, and London: University of California Press.
  3. Bhardwaj, P. (2018, July 6). Alibaba's new Hong Kong store combines the best parts of physical and online shopping - take a closer look. Business Insider. Retrieved from https://www.businessinsider.com/alibaba-fashion-ai-store-shopping-photos-2018-7
  4. Choo, S. (2001). Images of fashion fabric based on color and texture (Unpublished doctoral dissertation). Yonsei University, Seoul.
  5. Heo, J., & Park, Y. (2016). A study on the range and ranking of the color to be recognized as a brown. Proceedings of Korean Society of Color Studies, Fall Conference, Korea, 37-40.
  6. Jang, D., & Chun, J. (2007). Automated classification scheme generation using product attribute information. The KIPS Transactions: PartD, 14-D(5), 491-500. doi:10.3745/KIPSTD.2007.14-D.5.491
  7. Jeong, S., Vicent, M., Kim, J. K., Lee, B. W., Lee, S. Y., Lee, M. J., & Lee, W. J. (2017). The application of deep learning for image attribute recognition and classification. Proceedings of the Korea Software Congress 2017, Korea, 877-879.
  8. Jo, J., Lee, C., Lee, D., & Lim, H. (2018). An development of image retrieval model based on Image2Vec using GAN. Journal of Digital Convergence, 16(12), 301-307. doi:10.14400/JDC.2018.16.12.301
  9. Jung, J.-r., & Kim, M.-h. (2019). A study on AI-based design process innovation in fashion industry. Journal of Communication Design, 67, 140-153.
  10. Kim, C. E., & Lee, J. H. (2018). Trends of big data and artificial intelligence in the fashion industry. Journal of the Korean Society of Clothing and Textiles, 42(1), 148-158. doi:10.5850/JKSCT.2018.42.1.148
  11. Kim, C. S., & Pak, J. E. (2010). Observation of textile design from 2005-2009 fashion collection and development of pattern design - Focused on conversational design -. The Research Journal of the Costume Culture, 18(6), 1179-1193. doi:10.29049/rjcc.2010.18.6.1179
  12. Kim, J.-I. (2010). An implementation of automatic upper-lower clothes matching system using machine learning. Journal of Korea Multimedia Society, 13(3), 467-474.
  13. Kim, K.-H., Park, Y.-S., & Jung, E.-C. (2011). Development of classification system model of design information for integrated design research. Journal of Korean Society of Design Science, 24(3), 55-66.
  14. Kim, T. S. (2000). 분류의 이해 [Understanding of classification]. Seoul: Bibliographic Information Processing Society.
  15. Kim, Y. W. (2015). Relationship between sensibility expression factors of the fabrics and preferences of fashion image (Unpublished doctoral dissertation). Chungbuk National University, Cheongju.
  16. Ko, S. G., Yoo, S. J., & Kim, E. A. (2003). DB for the structural characteristics, images and sensibilities of fabrics-Effects of the structural characteristics on the texture images of woolen fabrics-. Journal of the Korean Society of Clothing and Textiles, 27(5), 533-544.
  17. Korean Agency for Technology and Standards. (2015). KS A 0011 Names of non - luminous object colours. Korean Standards & Certifications. Retrieved from https://standard.go.kr/KSCI/standardIntro/getStandardSearchView.do?menuId=919&topMenuId=502&upperMenuId=503&ksNo=KSA0011&tmprKsNo=KSA0011&reformNo=13
  18. Kwon, H.-j., & Kim, S.-j. (2012). Periodic expression research on retro focusing on the motif and structuralizing method of pattern design. Bulletin of Korean Society of Basic Design & Art, 13(4), 1-14.
  19. Lee, E. O. (2007). A study on the classification in Korean textile design. Journal of the Korean Society of Design Culture, 13(3), 271-280.
  20. Lee, K. H. (2012). Fashion design. Seoul: Kyungchoonsa.
  21. Lee, H., & Chang, Y. (2018). Development of an image tagging system based on crowdsourcing. Journal of the Korean Biblia Society for Library and Information Science, 29(3), 297-320. doi:10.14699/kbiblia.2018.29.3.297
  22. Lee, J. (2020). A study on the corresponding relationship between the KS A 0011 tone classification system and systematic color names (Unpublished master's thesis). Ewha Womans University, Seoul.
  23. Lee, J. E. (2018, December 2). 인공지능이 옷 만든다, 패션도'AI 열풍' [Artificial intelligence makes clothes, fashion is 'AI craze']. ChosunBiz. Retrieved from https://biz.chosun.com/site/data/html_dir/2018/11/30/2018113002436.html
  24. Lee, J.-H., & Kim, S.-W. (2011). Consumer-based apparel design elements and evaluation criteria-Focus on F/W outer-. Journal of the Korean Society of Clothing and Textiles, 35(9), 1015-1027. doi:10.5850/JKSCT.2011.35.9.1015
  25. Lee, K.-S., & Kim, H.-S. (1998). Fashion coordination. Seoul: Kyomunsa.
  26. Lee, S., & Lee, Y. (2018). Comparison of color recognition of Yellow Red, Green Yellow, Purple Blue and Red Purple according to the color name marking. Proceedings of Korean Society of Color Studies, Spring Conference, Korea, 134-137.
  27. Lim, K.-r. (2004). A study on classification system for using internet information resources on interior design. Journal of Korean Society of Design Science, 17(4), 79-88.
  28. Liu, Z., Luo, P., Qiu, S., Wang, X., & Tang, X. (2016). DeepFashion: Powering robust clothes recognition and retrieval with rich annotations. Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), USA, 1096-1104. doi:10.1109/CVPR.2016.124
  29. Nikkei Top Leader, & Nikkei Big Data. (Eds.). (2018). 일본 기업은 AI를 어떻게 활용하는가 [How Japanese companies utilize AI] (H. W. Shin, Trans.). Seoul: Paperroad. (Original work published 2017)
  30. Oh, H. S., & Park, H.-S. (2000). 아름다운 여성을 위한 패션코디 [Fashion coordination for beautiful women]. Seoul: Kyungchoonsa.
  31. Park, J. R. (2019). A study on the development of image based product search system using attribute classification (Unpublished master's thesis). Sejong University, Seoul.
  32. Park, M.-L., & Yong, H.-S. (2016). Objects classification in three level model using convolutional neural network. Proceedings of Korean Institute of Information Scientists and Engineers, Winter Conference, Korea, 131-133.
  33. Park, N., & Choi, Y. (2019). A form attribute classification system for fashion items. International Journal of Fashion Design, Technology and Education, 13(1), 11-19. doi:10.1080/17543266.2019.1664641
  34. Patki, R., & Suresha, S. (2016). Apparel classification using CNNs. CS231n: Convolutional Neural Networks for Visual Recognition. Retrieved from http://cs231n.stanford.edu/reports/2016/pdfs/286_Report.pdf
  35. Rhee. E.-Y. (2003). Theory of fashion design. Seoul: Kyomunsa.
  36. Seo, Y., & Shin, K.-s. (2018a). Business application of convolutional neural networks for apparel classification using runway image. Journal of Intelligence and Information Systems, 24(3), 1-19. doi:10.13088/jiis.2018.24.3.001
  37. Seo, Y., & Shin, K.-s. (2018b). Image classification of fine-grained fashion image based on style using pre-trained convolutional neural network. Proceedings of 2018 IEEE 3rd International Conference on Big Data Analysis (ICBDA), China, 387-390. doi:10.1109/ICBDA.2018.8367713
  38. Shin, C.-w., & Kim, C. (2018). Image category classification based on deep learning. Proceedings of Korea Computer Congress 2018, Korea, 95-97.
  39. Song, H. J., Kim, H. W., Chung, E., Oh, S., Lee, J. W., Kang, D., ... Lee, Y. K. (2019). Self-improving artificial intelligence technology. Electronics and Telecommunications Trends, 34(4), 43-54. doi:10.22648/ETRI.2019.J.340405
  40. Wang, X., & Zhang, T. (2011). Clothes search in consumer photos via color matching and attribute learning. Proceedings of the 19th ACM international conference on Multimedia, USA, 1353-1356. doi:10.1145/2072298.2072013
  41. Wong, C. W. K. (2018, May). Keynote Speech III: Challenges & opportunities of AI in fashion & textile industry (AI and computer vision). Paper presented at 2018 International Conference on Clothing & Textiles (ICCT), Seoul.