The Masking Machine (Feathered) © M Eifler
In late 2017, which feels like a digital lifetime ago, I spoke with author Alicia Eler about her debut book The Selfie Generation. Our conversation culminated with a question about the selfie as an aesthetic and communicative form that still knows no boundaries. Eler’s cogent observation came to mind when looking at the work of M Eifler, aka BlinkPopShift.
Eifler, a Bay Area multidisciplinary artist and XR (extended reality) designer pushes the boundaries of selfies through applied technology. Using a landmark detection algorithm, the artist repeatedly subjects these humble self-portraits to intense editing and manipulation until the algorithm can no longer detect their face.
The series Masking Machine (2018), recently featured in Recoding CripTech at SOMArts in San Francisco, activates questions about identity, self-representation, and mechanical intervention and authorship that date to photography’s early years. I spoke with Eifler about the motivation and method behind this unique series.
Roula Seikaly in conversation with M Eifler
The Masking Machine (Tropical) © M. Eiffler
Roula Seikaly: What motivated you to make this work?
Eifler: You know those brief moments in-between the unrelenting pain when you've had migraines 13 days in a row, but as long as you don't move the room has stopped spinning and the light doesn’t hurt your eyes? How do you make art then? Make a few self-portraits on your phone because those are the only two materials you have!
Seikaly: What does it mean to you to "make art embodied?"
Eifler: I am most interested in art when it is made not just for the eyes or ears but also the fingertips – when it's worn or performed or interacted with – when it's made from everyday materials we have preexisting physical/textural knowledge of when it's made while my body tries not to fall apart when it engages with the vast variety of pre-linguistic somatic knowledge that makes up being alive.
The Masking Machine (O mouth) © M. Eifler
Seikaly: On your website, you identify as an artist and XR designer and researcher. Does your work in and research of extended reality influence the breadth of your artistic practice?
Eifler: Actually, the breath in my artistic practice came before the codified position of design researcher. In grad school at CCA (California College of the Arts) a decade ago, I was told explicitly that much of my work was not art because of its use of and interest in technology. At that time the art world was not interested in reimaginings of user interfaces, projects questioning what is seeable through pixels, or unwinnable video games.
So, because I live in San Francisco, eventually the tech world came knocking at my door. Working in tech allowed me to push the boundaries of my experimentation by giving me access to technologies I would not otherwise have had. But even with that fuzzy boundary between my artwork and my experimental research my most far-flung work still continues to live in my artistic practice.
The Masking Machine (Slide) © M. Eifler
Seikaly: What is involved in working with the landmark detection algorithm? Could you briefly describe that process in layperson terms?
Eifler: In facial landmark detection the model is trained to take a picture of a face and give back a list of coordinates that are statistically likely to correspond to specific parts of that face. This list includes the inner and outer corners of the eyes, several spots along the curve of the eyelids, the nostrils and along the bridge of the nose. These points give you the overall shape of a face and allow you to overlay another image, like one of makeup, on it easily by matching up the points.
The model is created by a large group of humans labeling thousands of images of faces, often selfies scraped from the internet, with points indicating the coordinates of facial features. It's important to remember that the model is stored and reapplied human expertise, not auto-magic computer intelligence.
To show, step by step, how the Masking Machine images were created I made this video:
The Masking Machine (Red Chin, long) © M. Eifler
Seikaly: On your website, you aptly describe selfies as "the most quotidian, maligned, and rampant photography of our time." Do you think your work supports or defies that perception of selfies?
Eifler: Both and more. Masking Machine uses selfies in order to do 3 things.
Feel accessible. AI can often be given this aura of a big-fancy-special-difficult thing. Pairing it with a familiar-easy-small-simple thing makes the piece feel closer to and more open to viewers.
Reference Cindy Sherman's manual self reimagining photography
Reframe selfies as material, not product. We think of selfies as finished products in the same way we might consider a Cindy Sherman print on a wall in a gallery a finished product. But in practice selfies are a collage material from which online identities, social media platforms, and even landmark detection and face recognition algorithms are constructed and maintained.
The Masking Machine (Beside) © M. Eifler
Seikaly: Do you relate to these images as selfies? Are audiences looking at you, or someone else?
Eifler: Yes, the images are selfies. We are used to using the algorithms in photo editing apps to manipulate how selfies look. Those algorithms are visible, if subtle, in the images we use them to edit. Masking Machine uses a perhaps more extreme version of the same idea. The recursive nature of my process allows you to more clearly see both me and the algorithm in the same face.
Seikaly: How do you know when the image is "finished," if at all? Is there such a thing as too much manipulation using the landmark detection algorithm?
Eifler: The images are finished when it can no longer detect a face. This can take between 50 and 100 recursions.
The Masking Machine (Glasses) © M. Eifler
Seikaly: How does the still series relate to The Masking Machine as a performance piece?
Eifler: The performance piece, documentation video above takes the same technology and brings it to life. Using a custom wearable computer I can walk around any space wearing the still images now animated by my facial expressions. When seen through the screen hovering in front of my face I wear the images like an avatar, but unlike with the stills on a wall or images online I can reach out from behind the screen to shake hands and talk with viewers.