Presentation 2022-10-18
Examination of facial attractiveness features using geometric morphological analysis of and deep learning methods.
Takanori Sano, Hideaki Kawabata,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In psychology, various studies have been conducted on the features that constitute facial attractiveness. In recent years, studies that computationally model the relationship between facial features and perceived attractiveness have attracted much attention. The methods have used geometric morphometrics and deep learning methods. However, each method has been studied independently, and there has yet to be integrated research that considers both models. In this study, Therefore, we combined each method to examine the details of facial attractiveness features. The results showed that the eyebrow region tends to be related to attractiveness in male images regardless of which method is used, which is consistent with findings in psychology. This approach is expected to contribute to understanding highly universal facial attractiveness features and extend psychological results and engineering applications.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Facial attractiveness / Geometric morphometrics / Deep learning / CNN / Grad-CAM
Paper # HIP2022-52
Date of Issue 2022-10-10 (HIP)

Conference Information
Committee HIP
Conference Date 2022/10/17(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Kyoto Terrsa
Topics (in Japanese) (See Japanese page)
Topics (in English) Eye Movement (including Accommodation and Pupil), Spatial Perception (Depth Perception, Motion Perception, etc.), etc.
Chair Yuji Wada(Ritsumeikan Univ.)
Vice Chair Hiroyuki Umemoto(AIST) / Sachiko Kiyokawa(Nagoya Univ.)
Secretary Hiroyuki Umemoto(Kyushu Univ.) / Sachiko Kiyokawa(NICT)
Assistant Ippei Negishi(Kanazawa Inst. of Tech.) / Daisuke Tanaka(Tottori Univ.)

Paper Information
Registration To Technical Committee on Human Information Processing
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Examination of facial attractiveness features using geometric morphological analysis of and deep learning methods.
Sub Title (in English)
Keyword(1) Facial attractiveness
Keyword(2) Geometric morphometrics
Keyword(3) Deep learning
Keyword(4) CNN
Keyword(5) Grad-CAM
1st Author's Name Takanori Sano
1st Author's Affiliation Keio University(Keio Univ.)
2nd Author's Name Hideaki Kawabata
2nd Author's Affiliation Keio University(Keio Univ.)
Date 2022-10-18
Paper # HIP2022-52
Volume (vol) vol.122
Number (no) HIP-213
Page pp.pp.26-31(HIP),
#Pages 6
Date of Issue 2022-10-10 (HIP)