Artificial Intelligence extends Rembrandt’s Artistic DNA

Published in

ITNEXT

6 min readJan 30, 2018

Click here to share this article on LinkedIn »

For Plato, “forms”, such as beauty, are more real than any objects that imitate them. Though the forms are timeless and unchanging, physical things are in a constant change of existence. Where forms are unqualified perfection, physical things are qualified and conditioned.

Substance attribute theory, is about objecthood, positing that “A thing-in-itself” is a property-bearer that must be distinguished from the properties it bears. Plato’s recurring fascination was the distinction between ideal forms and everyday experience, and how it played out both for individuals and for societies.

Yet in this age of Artificial Intelligence we seem to have reverted to the idea that there is not “A thing-in-Itself”. There is nothing more than the attributes a thing bears, to be identified by deep neutral networks working to identify and match inputs to the AI application with attributes of things it finds in a database.

Paralleldots — Developed a deep Convolutional Neural Network (CNN) to classifying images into being those perceived as attractive or unattractive by humans. An images attractiveness is determined based on the potential of an image to “go viral on the Internet.” The Paralleldots “viral detection API” uses Google’s AVA2 Dataset to train it’s algorithms to find attributes of images that have gone viral on the Internet. While this is not the same as the beauty attributed to an image because of objective characteristics, it is quite similar to Immanuel Kant’s condition of “universality” which he used as one measure of beauty.

The Next Rembrandt — ING, a Dutch multinational banking group, was looking for a way to innovate and stand out amongst its competitors. Because ING is a longtime supporter of Dutch arts and culture — a world that’s steeped in tradition — “art” became the natural playground for the brand’s venture into innovation.

Made solely from data derived from 346 known paintings by Rembrandt, “The Next Rembrandt” was created from a deep learning, 18-month analysis of attributes in a dataset of the artist’s works.

A facial-recognition algorithm learned Rembrandt’s techniques; pixel data helped the computer mimic brushstrokes; and an advanced 3D printer brought the painting to life using 13 layers of ink. The portrait consists of 148 million pixels and is based on 168,263 attributes of data from Rembrandt’s portfolio.

Thread Genius - is an AI application purchased by Sotheby’s to match art to collectors. Thread Genius, is a company founded by former Spotify engineers that uses artificial intelligence to identify collector tastes and offer recommendations.

Evan Beard, the executive of U.S. Trust National Art Services says Sotheby’s is looking at horizontal revenue streams because the vertical stream of auctioning things is a very challenging one,”
“We know our clients not only buy fine art, but they buy decorative art, wine, jewelry and watches,” says Jennifer Deason, Sotheby’s head of strategy and corporate development. “If we have the capability to recommend other items they may like, there’s a real power to that recommendation engine.” If a collector likes 20th century Mies van der Rohe chairs, they may be “more apt to buy a certain kind of time period furniture,” Deason says.

Beauty.ai — is an AI application that allows individuals and others to submit photographs they consider to be beautiful and the app is trained to evaluate attributes of beauty like:

Face Symmetry
Skin Color
Wrinkles
Gender
Age Group
Ethnicity

to determine the beauty “winners”. According to the Beauty.ai team, their challenge is “to find effective ways to slow down aging and help people look healthy and beautiful.”

Following are the Beauty.ai winners in the Women’s 18–29 age group along with their scores on various attributes of “beauty”.

As can be noted on the “Beauty.ai” winners sheet a “Symmetry Master Score” is calculated for each contestant. Researchers from the University of Stockholm have reported in “Nature” ( Nature volume 372, pages 169–172 (10 November 1994 ) “humans and certain other species find symmetrical patterns more attractive than asymmetrical ones.”

The researchers show the existence of sensory biases for symmetry may have been exploited independently by natural selection acting on biological signals and by human artistic innovation. This may account for the observed convergence on symmetrical forms in nature and decorative art. From Leonardo da Vinci to Le Corbusier, the golden ratio is believed to have guided the symmetry applied by artists and architects over the centuries.

Leonardo is thought to have used the golden ratio, a geometric proportion regarded as the key to creating aesthetically pleasing art, when painting the Mona Lisa. The Dutch painter Mondrian used it in his abstract compositions, as did Salvador Dali in his masterpiece The Sacrament of the Last Supper.

In “The Sacrament of the Last Supper,” Salvador Dali framed his painting in a golden rectangle. Following Da Vinci’s lead, Dali positioned the table exactly at the golden section of the height of this painting. He positioned the two disciples at Christ‘s side at the golden sections of the width of the composition. In addition, the windows in the background are formed by a large dodecahedron. Dodecahedrons consist of 12 pentagons, which exhibit phi relationships in their proportions.

Now Adrian Bejan, a professor of mechanical engineering at Duke University, in Durham, North Carolina, believes he has discovered the reason why the golden ratio pleases the eye. According professor Bejan the human eye is capable of “interpreting an image” featuring the golden ratio faster than any other.

The closer a given subject is to the ratio, the more beautiful it is generally considered to be. This means that in a ‘beautiful’ face, like the adjacent one, the distance from the middle of the nose is 1.618’th of the width of the eye. Another example is the distance from the chin to the pupil is 1,618’th of the distance from the pupil to the hairline. In the body, the distance from the sole of the foot to the waist is 1,618’th of the distance from the waist to the crown.

In a very recent study Nathan Kondamuri, of the Art Institute of MIT, took measurements using two sculptures found in the Hayden Memorial Library at MIT. He also conducted a survey asking subjects to compare the two sculptures and rate them on a scale of 1–10 on their beauty. Upon comparing the measurements and results from this survey, Kondamuri concluded “there is a direct correlation between the golden ratio and aesthetic beauty”.

Leonardo da Vinci’s Mona Lisa had a number of golden rectangles throughout. By drawing a rectangle around her face, we can see she is indeed golden. If we divide that rectangle with a line drawn across her eyes, we get another golden rectangle, meaning that the proportion of her head length to her eyes is golden. There are other golden rectangles that can be drawn on the rest of her body like from the her neck to the top of her hands. So while Mona Lisa may have been seen as beautiful, using an objective standard like the golden ratio, an analysis of her “attributes” may not extend her beauty using 21st century Artificial Intelligence and the “big data” of which da Vinci’s enormous body of work provides just some of the attributes. In fact when virality scores for the 10 most famous Paintings In the world were calculated using the Paralleldots AI application da Vinci’s Mona Lisa ranked 10th out of 10 with a virality score of only 9% compared to a score of 95% for Munch’s “the Scream”.

_________________________________________________________________

Notes:

Thanks for reading! My work is entirely reader-funded so if you enjoyed this piece please help us spread the word on using AI to share beauty and contribute on Paypal,

Artificial Intelligence extends Rembrandt’s Artistic DNA

Written by william smith