Estimation of the Attractiveness of Food Photography Based on Image Features

Abstract

Nonmembers, Keisuke DOMAN † †b) , Yasutomo KAWANISHI † † †c) , Takatsugu HIRAYAMA † † † †d) , Members, Ichiro IDE † † †e) , Senior Member, Daisuke DEGUCHI † † † † †f) , Member, and Hiroshi MURASE † † †g) , Fellow SUMMARY We introduce a method to estimate the attractiveness of a food photo. It extracts image features focusing on the appearances of 1) the entire food, and 2) the main ingredients. To estimate the attractiveness of an arbitrary food photo, these features are integrated in a regression scheme. We also constructed and released a food image dataset composed of images of ten food categories taken from 36 angles and accompanied with attractiveness values. Evaluation results showed the effectiveness of integrating the two kinds of image features.

Factors influencing the visual deliciousness / eye-appeal of food

“…As such, we have not looked at ways to make the food look more sustainable. The behavioural and cognitive neuroscience research that has been published to date clearly demonstrates that the appeal of a given portion of food can vary as a function of how exactly it is presented: Everything from the viewing angle at which the food is shown (e.g., Michel et al, 2015;Takahashi, Hattori, Doman, Kawanishi, Hirayama, Ide, Deguchi, & Murase, 2019; though see also Sakay Rodriguez et al, in press) through to the layout/orientation of the food on the plate or meal tray/basket/platter (e.g., Michel, Velasco, Gatti, & Spence, 2014;Saner, 2015), and from the background against which the food happens to be presented through to the level of lighting in the food image (Howell & Schifferstein, 2019;Spence, 2018a), or even in the environment in which that food image is being viewed (e.g., Papies, van Stekelenburg, Smeets, Zandstra, & Dijksterhuis, 2022). In fact, according to the research that has been published to date, it all appears to make a difference.…”

“…Such a dissociation may be hard to square with attempts to use image analysis techniques in order to try and computationally analyse/predict the deliciousness of food images (see Takahashi et al, 2019). At the same time, however, black and white, rather than colour, images of food appear to be rated as significantly less appealing despite presumably portraying exactly the same food (Del Gatto, Indraccolo, Imperatori, & Brunetti, 2021; see also Sato, 2021).…”

Perancangan Media Fotografi Makanan sebagai Rekomendasi Jajanan Jalanan di Batam menggunakan Metode IDI

“…Jadi, pemotretan gambar harus diambil dengan profesional agar hasilnya dapat membuat publik menjadi tertarik. Untuk menghasilkan gambar yang menarik, dibutuhkan dukungan fotografi yang dapat mempertimbangkan daya tarik makanan yaitu sudut kamera terbaik untuk masing-masing hidangan [7]. Michael et al melaporkan bahwa ada sudut kamera dimana sebuah makanan terlihat sangat menarik.…”

“…Penelitian oleh [7] adalah penelitian yang bertujuan untuk mengukur estimasi daya tarik fotografi makanan berdasarkan fitur gambar yang diberikan. Penelitian ini menggunakan metode Thurstone's paired comparison untuk menentukan daya tarik fotografi makanan.…”