Journal of Internet of Things and Convergence (사물인터넷융복합논문지)

The Korea Internet of Things Society (한국사물인터넷학회)
권호 표지
DOI QR코드

DOI QR Code

Implementation of Hair Style Recommendation System Based on Big data and Deepfakes

빅데이터와 딥페이크 기반의 헤어스타일 추천 시스템 구현

  • Tae-Kook Kim (School of Computer and Artificial Intelligence Engineering, Pukyong National University)
  • 김태국 (부경대학교 컴퓨터.인공지능공학부)
  • Received : 2023.05.18
  • Accepted : 2023.06.14
  • Published : 2023.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we investigated the implementation of a hairstyle recommendation system based on big data and deepfake technology. The proposed hairstyle recommendation system recognizes the facial shapes based on the user's photo (image). Facial shapes are classified into oval, round, and square shapes, and hairstyles that suit each facial shape are synthesized using deepfake technology and provided as videos. Hairstyles are recommended based on big data by applying the latest trends and styles that suit the facial shape. With the image segmentation map and the Motion Supervised Co-Part Segmentation algorithm, it is possible to synthesize elements between images belonging to the same category (such as hair, face, etc.). Next, the synthesized image with the hairstyle and a pre-defined video are applied to the Motion Representations for Articulated Animation algorithm to generate a video animation. The proposed system is expected to be used in various aspects of the beauty industry, including virtual fitting and other related areas. In future research, we plan to study the development of a smart mirror that recommends hairstyles and incorporates features such as Internet of Things (IoT) functionality.

본 논문에서는 빅데이터와 딥페이크 기반의 헤어스타일 추천 시스템 구현에 관해 연구하였다. 제안한 헤어스타일 추천 시스템은 사용자의 사진(이미지)을 바탕으로 얼굴형을 인식한다. 얼굴형은 타원형, 둥근형, 장방형으로 구분하며, 얼굴형에 잘 어울리는 헤어스타일을 딥페이크를 통해 합성하여 동영상으로 제공한다. 헤어스타일은 빅데이터를 바탕으로 최신 트랜드(trend)와 얼굴형에 어울리는 스타일을 적용하여 추천한다. 이미지의 분할 맵과 Motion supervised Co-Part Segmentation 알고리즘으로 같은 카테고리(머리, 얼굴 등)를 가지는 이미지들 간 요소를 합성할 수 있다. 다음으로 헤어스타일이 합성된 이미지와 미리 지정해둔 동영상을 Motion Representations for Articulated Animation 알고리즘에 적용하여 동영상 애니메이션을 생성한다. 제안한 시스템은 가상 피팅 등 전반적인 미용산업에 활용될 수 있을 것으로 기대한다. 향후 연구에서는 거울에 사물인터넷 기능 등을 적용하여 헤어스타일등을 추천해주는 스마트 거울을 연구할 예정이다.

Keywords

Acknowledgement

이 성과는 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구임(RS-2023-00242528).

References

  1. B.H.Kang, "A study on the interest and image of men's hairstyles," Nambu university, Doctoral thesis, 2022.
  2. H.M.Kim, J.S.Lee, "A Study on the Male Hair Cut Section Division," Journal of the Korean Society of Cosmetology," Vol.23, No.4, pp.811-823, 2017.
  3. Y.S.Ku, Y.J.Lee, and T.G.Choo, "A Study on Appearance Management Behavior of Maile Cosumers(I)," The Society of Fashion and Textile Industry, Vol.12, No.4, pp.459-466, 2010. https://doi.org/10.5805/KSCI.2010.12.4.459
  4. EJ.Im, M.W.Kim, "The Correlation between Recognition of Hair style and Body Image of Men in his Twenties," Journal of the Korean Society of Design Culture, Vol.20, No.2, pp.523-533, 2014.
  5. J.Y.Son, "A Study on the Perception of Image According to Face and Hair Style," Konkuk university, Doctoral thesis, 2011.
  6. H.M.Noh, H.W.Joo, Y.S.Moon, K.S.Kong, "Smart Mirror to support Hair Styling," The Journal of the Institute of Internet, Broadcasting and Communication, Vol.20, No.1, pp.127-133, 2020.
  7. S.B.Kang, S.H,Kwon, Y.H.Kim, S.I.Lee, Y.O.Han, "Smart Mirror for Styling," The Journal of the Institute of Internet, Broadcasting and Communication, Vol.16, No.6, pp.1271-1278, 2021.
  8. OpenCV, https://opencv.org/
  9. OpenCV Face Recognition, https://opencv.org/opencv-face-recognition/
  10. Dlib C++ Library, http://dlib.net/
  11. Dlib Face Detector, http://dlib.net/face_detector.py.html/
  12. A.Siarohin, S.Roy, S.Lathuiliere, S.Tulyakov, E.Ricci, N.Sebe, "Motion-supervised Co-Part Segmentation", ICPR 2021.
  13. A.Siarohin, O.J.Woodford, J.Ren, M.Chai, S.Tulyakov, "Motion Representations for Articulated Animation", CVPR 2021.
  14. J.H.Kim, "An Classification of Face, Ears, Eyes, Mouth and Nose Type of Women in Twenty", The Catholic university of Korea, Doctoral thesis, 2008.
  15. M.Y.Song, O.L.Park, "A Study on Women's Face Types Classification and Shape Differences," Journal of Fashion Business, Vol.8, No.1, pp.76-90, 2004.