Research on how to present images showing uneven foundation application that do not cause unfavorable impressions to the viewer

Miho Kajita, Satoshi Nakamura, Takayuki Itoh

Session: Tangibles and AI - Wednesday 2022-11-30 14:00-15:30 AEDT
Late Breaking Work and Case Studies Track
In-Person Presentation


Foundation is expected to have a covering effect that hides skin imperfections such as pores and uneven coloring when it is applied evenly. However, since the foundation itself is close to the skin color, people cannot tell easily where they have applied it. Therefore, we proposed a system to detect and visualize uneven foundation application, and have succeeded in binary classification of bare skin and foundation skin. Here, visualizing the unevenness of foundation on the face may create a negative impression for users. Therefore, we studied a method to visualize the uneven application of foundation on the face, taking into consideration factors other than ease of understanding. As a result, we found that the most positive evaluation was a method that made it relatively easy to recognize where foundation had been applied unevenly and gives the impression of “bright, warm, active and radiant”.