Photo Corners

A   S C R A P B O O K   O F   S O L U T I O N S   F O R   T H E   P H O T O G R A P H E R

Enhancing the enjoyment of taking pictures with news that matters, features that entertain and images that delight. Published frequently.

Yahoo: 'Look Sharp!' To Be Beautiful. But... Share This on LinkedIn   Share This on Google   Tweet This   Forward This

5 February 2015

The other day the Physics arXiv Blog reported, "Scientists at Yahoo Labs have developed a machine learning algorithm that distinguishes beautiful portraits from the not-so."

The blog story is based on The Beauty of Capturing Faces: Rating the Quality of Digital Portraits by Miriam Redi, Nikhil Rasiwasia, Gaurav Aggarwal, Alejandro Jaimes. The short version follows:

Digital portrait photographs are everywhere, and while the number of face pictures keeps growing, not much work has been done to on automatic portrait beauty assessment. In this paper, we design a specific framework to automatically evaluate the beauty of digital portraits. To this end, we procure a large dataset of face images annotated not only with aesthetic scores but also with information about the traits of the subject portrayed. We design a set of visual features based on portrait photography literature, and extensively analyze their relation with portrait beauty, exposing interesting findings about what makes a portrait beautiful. We find that the beauty of a portrait is linked to its artistic value, and independent from age, race and gender of the subject. We also show that a classifier trained with our features to separate beautiful portraits from non-beautiful portraits outperforms generic aesthetic classifiers.

According to the blog report, Redi's team at Yahoo Labs in Barcelona began with a set of 10,000 photographic portraits that had been rated for beauty by humans on a scale from 1 to 10 with additional notes about the images.

We suspect what the researchers have developed is not so much a way to detect human beauty but a way to detect a professional studio portrait.

Face recognition software determined "the age, sex and race of the main subject" in each portrait, mapping the "relative coordinates of the eyes, nose and mouth."

The images were then ranked by software according to composition, exposure and lighting taking into account the arrangement of objects within the image, the distribution of lighting and the sharpness of the picture. Also factored in were the exposure, contrast and JPEG compression plus with the contrast between the face and the background.

Finally, they used the human notes to classify the machine findings as positive or negative.

And what they got from all this, the report says, is a machine algorithm for recognizing human beauty.


The team found that "race, gender, and age are largely uncorrelated with photographic beauty." Instead, "aesthetic score is related to sharpness of facial landmarks, image contrast, exposure, homogeneity, illumination pattern, uniqueness and originality."

And, perhaps most revealing, the story reports:

Indeed, the single most important factor is the sharpness of the image. But other important factors include the contrast between the face and background. Curiously, exposure quality is negatively correlated with beauty suggesting that photographers can create beautiful images by playing with under and overexposed images.

All leading to the claim by the Yahoo team that "we built a classifier that is able to successfully distinguish between beautiful and non-beautiful portraits."


There's a little disconnect between what the researchers said and what the blog (and a few other stories on this) reports.

The researchers talk about "photographic beauty" while the report talks about "human beauty." Sharpness, for example, distinguishes beautiful portraits not "age."

And that isn't the only formal photographic quality that scored above mere physical traits (with which human beauty concerns itself). Wny would the researchers be looking at exposure, lighting, contrast and compression?

We suspect what the researchers have developed is not so much a way to detect human beauty but a way to detect a professional studio portrait. You know, one in which the photographer used a tripod to get that tell-tale sharpness.

A beautiful portrait does not require a beautiful face. A beautiful face does not guarantee a beautiful photo.

Ask any portrait photographer.

BackBack to Photo Corners