The Art of Feeling

Turning feelings into digital paintings

Mindswarms was commissioned by the Saatchi & Saatchi Wellness agency as part of their new identity. We developed a system to produce digital paintings from emotions. We asked each of their employees to think of an emotional memory then used an EEG headset to record their brainwaves.

We then used emotion recognition algorithms to determine what they were feeling and fed this data into a generative algorithm to create swarm paintings that represented each individual's state of mind at the time of recording.

These paintings have become a fundamental part of Saatchi & Saatchi Wellness' identity and exhibited in a gallery in London.

To produce complex and unique visualisations of the emotional memories of each employee we were inspired by bird swarms and used generative techniques. Swarm behaviour is the collective movement effect of a number of individual 'agents' that follow simple rules to guide their behaviour.

Each agent was represented as a tiny brush that moved on a canvas leaving an ink trail behind it. Emotional data extracted from the raw brainwaves was fed into the swarm visualisation system which resulted in complex, non-repeating patterns of colour.

Press

Emotional paintings in exhibition

The paintings of each employee's emotional memory were printed in high resolution, framed and exhibited in the Truman brewery gallery space in East London. The space was filled with emotional snapshots captured in paper such as the "The birth of my son" and "Going shark cage diving in South Africa".

Personalised business cards

Each employee also had their emotional painting printed on the back their business cards, along with the thought that inspired that emotion.

Turning feelings on twitter into live visualisations

In Saatchi & Saatchi Wellness' homepage we also created a real time version of the visualisation. The site uses webGL to create a highly optimised moving version of the swarm visualisations. Instead of brain data we turned to Twitter. The website gathers live tweets and processes them using a machine learning sentiment analysis engine. It then uses the quantified emotional output to create a moving digital painting that reflects the emotional state of the world in present time.

The website can be viewed live at saatchiwellness.co.uk

Special thanks: Dr. Mick Grierson, Goldsmiths, University of London