Chelsi Cocking

Photorythms: a computational art-based inquiry of portrait photography

Photorythms investigates whether computational methods such as facial detection, computer vision, and generative forms can be utilized to create more expressive and artistic works of portraiture and the face. Giving a new take on portrait photography and new life to images through computation.

  • Can portrait photos and photos of people be more expressive than they are today?
  • Can computational methods assist this?

Photorythms has also been implemented as a public interactive art installation that invites participants to sit and have their faces detected and augmented live in real-time. Learn more through the video below and on the Photorythms: Interactive project page.


Inspired by the work of Berenice Abbott, who pioneered work around the photography and documentation of physics at MIT, as well as Harold "Doc" Edgerton, an electrical engineer and notable MIT professor who utilized the electronic stroboscope to capture motion in images that, at the time, made the invisible visible. 

Berenice Abbott

"Renowned for her early to mid-century photography in Paris and New York, Abbott also spent time at MIT during the late 1950s when she was hired to create new photographic images for the teaching of physics. 

Berenice Abbott spent two years at MIT creating photographs that memorably document the principles of physical science—mechanics, electromagnetism, and waves. She often developed innovative techniques for capturing scientific phenomena, including one for very detailed, close-in photography that she called Super Sight."

– MIT Museum, Berenice Abbott: Photography and Science: An Essential Unity

Harold "Doc" Edgerton

"Born in Fremont, Nebraska, Harold “Doc” Edgerton (1903–1990) began his graduate studies at MIT in 1926. He became a professor of electrical engineering at MIT in 1934. In 1966, he was named Institute Professor, MIT's highest honor.

With his development of the electronic stroboscope, Edgerton set into motion a lifelong course of innovation centered on a single idea—making the invisible visible. An inveterate problem-solver, Edgerton succeeded in photographing phenomena that were too bright or too dim or moved too quickly or too slowly to be captured with traditional photography."

– MIT Museum, Flashed of Inspiration: The Work of Harold Edgerton

Also inspired by today's contemporary artists who use artwork to tackle the subjects of portraiture, the body, and the face in different ways from unique lenses and perspectives. 

Get the data

Created in XCode, I utilized OpenFrameworks, an open-source C++ toolkit for creative coding, to build the Photorythms project. Facial landmark points were detected using the OpenFramworks addons ofxFaceTracker2 created by Jonas Jongejan and OpenFramworks addons ofxCV created by Kyle McDonald. The project was structured so that I could code multiple computational sketches based on the facial landmark points detected on an image (or multiple images) to explore different visual ideas and results. Depending on what I wanted to do with the sketch, I then created various texture-based shapes filled based on the image. I call these customized textures "face shapes".

Use the Data

The goal was to create a series of computational sketches that explored different generative forms and visual ideas based on facial landmark points.

I would usually start by hand sketching an idea I want to explore. 

From there, my personal goal is to enjoy the process of creating, experimenting, and having fun. Creating new forms of portraiture and new ways of approaching photography and the face through code. Utilizing "face shapes" to see if my original ideas can be realized and if new ideas are sparked during the making process. This open experimentation allowed me to breathe new life into an image through exploration and computation. 

Various Outputs

Grid Study

Expanded Face

Scaling Circles