Presentation | 2022-09-15 Comparison of results for different datasets of facial attractiveness features extracted using gradient-weighted class activation mapping Takanori Sano, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In recent years, research using deep learning to analyze faces to identify the features that make up facial attractiveness has attracted much attention. However, the differences in results depending on the dataset used for analysis have not been studied in detail. In this study, I constructed convolutional neural network models for each dataset using two datasets and visualized the features important for prediction using the gradient-weighted class activation mapping method. The results showed that, for both datasets, the eye region tended to be activated in highly attractive female images, which is consistent with findings in psychology. This approach is expected to contribute to understanding highly universal features in facial attractiveness and contribute to various engineering applications in addition to extending psychological findings. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Facial attractiveness / Psychology / Grad-CAM |
Paper # | AI2022-24 |
Date of Issue | 2022-09-08 (AI) |
Conference Information | |
Committee | AI |
---|---|
Conference Date | 2022/9/15(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Yuichi Sei(Univ. of Electro-Comm.) |
Vice Chair | Yuko Sakurai(AIST) / Tadachika Ozono(Nagoya Inst. of Tech.) |
Secretary | Yuko Sakurai(Tokyo Univ. of Agriculture and Technology) / Tadachika Ozono(Toho Univ.) |
Assistant | Kazutaka Matsuzaki(Chuo Univ.) |
Paper Information | |
Registration To | Technical Committee on Artificial Intelligence and Knowledge-Based Processing |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Comparison of results for different datasets of facial attractiveness features extracted using gradient-weighted class activation mapping |
Sub Title (in English) | |
Keyword(1) | Facial attractiveness |
Keyword(2) | Psychology |
Keyword(3) | Grad-CAM |
1st Author's Name | Takanori Sano |
1st Author's Affiliation | Keio University(Keio Univ.) |
Date | 2022-09-15 |
Paper # | AI2022-24 |
Volume (vol) | vol.122 |
Number (no) | AI-186 |
Page | pp.pp.37-41(AI), |
#Pages | 5 |
Date of Issue | 2022-09-08 (AI) |