Artificial Intelligence Used to Look At Great Art Through the Lens of Today's News

Wednesday, October 26, 2016

Artificial Intelligence Used to Look At Great Art Through the Lens of Today's News

Artificial Intelligence

The award-winning piece of software, 'Recognition' uses artificial intelligence to compare up-to-the-minute photojournalism with British art from the Tate collection. For the next few months, the software will be used to create a constantly evolving display at the museum, and online.

'Recognition', winner of the Tate Britain's IK Prize 2016 for digital innovation, is an artificial intelligence program that compares up-to-the-minute photojournalism with British art from the Tate collection. The project was developed by a team at Fabrica, a communication research center in Treviso, Italy. The algorithm continuously screens about 1,000 news photographs a day, supplied by Reuters, and tries to match them with 30,000 British artworks in the Tate’s museum database, based on similarities in faces, objects, theme and composition.

Meredith Frampton Portrait of Sir George Frampton 1925, Tate © Estate of Meredith Frampton REUTERS
"Art is not so simple; art is uncharted space for AI."

Comparisons between artistic works and other material are made by the software program and are for the purpose of stimulating debate about art, expression and representation.

The IK Prize is presented annually by Tate for an idea that uses digital technology to innovate the way people discover, explore and enjoy British art in the Tate collection.

The 2016 prize, challenged digital creatives to use artificial intelligence to explore, investigate or ‘understand’ British art in the Tate collection. Selected as the winner for Recognition, Fabrica received a cash prize, production budget and support from Tate and Microsoft.

The Fabrica team collaborated with artificial intelligence specialists and web developers Jolibrain to turn their idea into reality. JoliBrain are artificial intelligence specialists based in Toulouse, France and editors of the DeepDetect deep learning API and server used in a variety of industries. The team are experts across the field of AI and its applications in a range of industries, from image recognition to NLP applications and cyber security.

“A.I. was chosen as the theme this year because getting machines to do what humans can do is one of the most exciting frontiers in technology,” Tony Guillan, a multimedia producer for Tate who manages the prize, told the New York Times. “Is there anything more human than looking at art?”

“We asked, ‘What if we could link our everyday lives to the collection to illuminate similarities between the present and the past?’” said Angelo Semeraro, a member of the Fabrica team.

Over three months, Recognition will create an ever-expanding virtual gallery at the museum by searching through Tate’s collection of British art and archive material online, comparing artworks  twenty-four hours a day.

"Making unforeseen comparisons across history, geography and culture, the result will be a time capsule of the world represented in diverse types of images, past and present," states the museum.

Recognition incorporates multiple artificial intelligence technologies, including computer vision capabilities such as object recognition, facial recognition, color and composition analysis; and natural language processing of text associated with images, allowing it to analyze context and subject matter and produce written descriptions of image comparisons.

Recognition is an artificial intelligence comparing up-to-the-minute photojournalism with British art from the Tate collection.


Recognition has four different ways of looking at an image:
  • Object recognition is a process for identifying specific objects. Its algorithms rely on matching, learning, or pattern recognition using appearance-based or feature-based analysis. 
  • Facial recognition is a process for identifying human faces. In addition to locating the human faces in an image, it determines the age, gender, and emotional state of each subject it finds. 
  • Composition recognition is a process for identifying prominent shapes and structures, visual layout, and colors. 
  • Context recognition is a process which analyses the titles, dates, tags, and descriptions associated with each image. By reading this text, it's also how recognition learns how to write a caption for each match. 
Images with close similarity in these four categories are selected as a match, and displayed in Recognition's gallery.

Related articles
The display at Tate Britain accompanies the online project offering visitors the chance to interrupt the machine’s selection process. The results of this experiment – to see if an artificial intelligence can learn from the many personal responses humans have when looking at images – will be presented on the virtual gallery site at the end of the project.

You can also view the “Recognition” process online: Images in the database rotate past a photo and are given scores, according to the four variables. When a match is made, the pairing is saved in an online gallery and displayed at Tate Britain. Since it began at the beginning of September, the program has been twinning images at a rate of one to three an hour. By the time it ends on Nov. 27, “we expect 2,000 to 3,000 matches,” Guillan said.

“Art is not so simple; art is uncharted space for AI,” said Eric Horvitz, director of the Microsoft Research Lab. Mr. Horvitz, who served on the prize’s judging panel, said that was one reason Microsoft collaborated with the Tate.

Visit the virtual gallery as it evolves online and help re-train the algorithm in an installation at Tate Britain.

SOURCE  Tate Britain

By  33rd SquareEmbed


Post a Comment