Los Angeles-based Adam Ferriss’ 2018 art installation, Likeness, takes one audience member’s face and transforms it into another. It “reimagines face-filtering and remolds people’s appearances in real time.” [1] Likeness, curated by Alex Czetwertynski, another digital artist and curator, was produced for Google IO 2018 in the Museum of Developer Art and commissioned by UCLA Conditional Studio.
It is similar to Tim Hawkinson’s Emoter (2000), which features the same concept of manipulating different facial characteristics to create an image of a human face, or amalgamated human-like attributes. Another remotely similar installation is Catherine Ikam’s and Louis Fléri’s Le Messager (‘The Messenger, 1995), an interactive 3D digital scan of a head fitted with sensors that tracked the movements of the audience, following the people around with its eyes. Ferriss’s installation featured a camera that would locate key focal points on the subject in front of the lens, prompting the AI on the LED screen to generate an image from the given face.
Ferriss began his installation by presenting the computer two sets of images: generic photographs of people, and what Ferriss calls “label maps” that “identify and highlight the different facial features, like eyes, eyebrows, jawline, mouth, and nose, present in the photographs.” [2] The computer would then juxtapose the two sets of images and make connections between the labels and photos in terms of facial focal points; Ferriss states that this is called training, and can take hours or days at a time. His computer was trained for the span of a week for thirty to forty hours, “using the pix2pixHD architecture on the Helen Dataset to generate faces from the audience in realtime.” [3] Once training had been finalized, Ferriss presented a “label map” to the computer, which modifies the facial construction of the subject. The final result could have changed any characteristic, including gender or skin color, or layered attributes on top of one another, as pictured in the photographs. Likeness was interactive, and allowed people to stand in front of a camera and pose for the AI face-generator, installed on an eight by ten-foot-high LED screen/wall.
All characteristics considered, a key takeaway from this installation is that no matter our differences, we are all human, deep down. Regardless of our cultural backgrounds, upbringings, and appearances, we all share the same element of being human and feeling emotions. For Ferriss himself, the purpose of the installation was “to shine a light on the playful aspect of AI; to highlight how technological advancements shouldn’t be seen as either apocalyptic or the solution to all world problems.” [4]
Ferris began coding seven years ago during his undergraduate program in photography at Maryland Institute College of Art, and had a large interest in learning to manipulate photographs. After taking a new-media class and having learned JavaScript and rudimentary coding, he began working on manipulation and his curiosity led him to where he is today. His “driving principle is to find new ways to interpret, distort, and redraw images and photos.” [5]
The official website for Ferriss’s installation is viewable here, along with an interactive GIF and more photographs of the installation: https://amf.fyi/Likeness
[1] [2] www.itsnicethat.com/articles/adam-ferriss-likeness-digital-190618
[3] https://amf.fyi/Likeness
[4] https://www.itsnicethat.com/articles/adam-ferriss-likeness-digital-190618
[5] dtimes.com/art/artists-use-code-to-create-mind-bending-digital-art/