Using computer vision techniques on Instagram to link users’ personalities and genders to the features of their photos: An exploratory study

Instagram and other photo-based social networking sites have emerged as a popular medium. Previous studies mainly focused on social media texts, but the current study deals with the relationships between the characteristics of Instagram users and the features of their photos. The Big Five personality traits and gender were employed as the variables for user characteristics. Content category, the number of faces, the emotions on the faces, and the pixel derived features were employed as the variables for photo characteristics. An online survey of 179 university students was conducted to measure user characteristics, and 25,394 photos in total were downloaded and analyzed from the respondents’ Instagram accounts. Results suggested that content category is associated with extraversion and gender of users. The number of faces is associated with extraversion, agreeableness, and openness of users. Extraversion, agreeableness, and openness of users were partly associated with emotions expressed on the faces in their photos. Correlations were observed among some pixel features and extraversion, agreeableness, conscientiousness, and gender of users. It was also observed that the Big Five personality traits, except for gender, can be predicted by above variables. Implications and limitations are discussed and suggestions for future research are suggested.