In this paper, we outline an image-based methodology which uses a large number of photos of a user in order to obtain their skin-tone. The motivation here is that by having a large number of images, such as those readily available on social networks, it will be possible to compensate for different lighting and photo conditions thereby obtaining a reasonable accurate measure of the true skin-tone. As shown in Figure 1, our methodology has been translated into a mobile application which finds a user’s skin tone, allows them to make corrections (for data collection purposes), and recommends the best matching foundation shades.