How Visual Aesthetics and Calorie Density Predict Food Image Popularity on Instagram: A Computer Vision Analysis

ABSTRACTSocial media have become an important source where people are exposed to visual representations of foods. This study aims to understand what content factors contribute to the popularity of food images on Instagram. We collected 53,894 images from 90 popular food influencer accounts on Instagram over two years. Applying computer vision methods, we investigated the effects of visual aesthetics and calorie density of foods on audience engagement (i.e. likes, comments) as well as if the effects of visual aesthetics varied by calorie density. Our results showed that both visual aesthetics and calorie density were important predictors of image popularity. The use of arousing, warm colors such as red, orange, and yellow, feature complexity, and repetition predicted higher likes, whereas brightness, colorfulness, and compositional complexity acted reversely. A similar pattern was observed for comments. The calorie density of foods in images positively predicted likes and comments. Also, the effects of visual aesthetics varied by calorie content and were more pronounced for low-calorie images. Health practitioners who plan to harness the power of social media to encourage certain dietary behaviors should take visual aesthetics into account when designing persuasive messages and campaigns. Code availability statementThe Python scripts to conduct computer vision analysis described in this manuscript are available at https://github.com/yilangpeng/food-image-instagram. The Python Package Athec (https://github.com/yilangpeng/Athec) was used to conduct the analysis of aesthetic features (e.g., brightness, color percentages). A detailed description can be found in Peng (Citation2022).Disclosure statementNo potential conflict of interest was reported by the author(s).Supplementary dataSupplemental data for this article can be accessed online at https://doi.org/10.1080/10410236.2023.2175635.Additional informationFundingThe author(s) reported there is no funding associated with the work featured in this article.