







The original concept of pixel portraits was done in Photoshop, where I have precise control in placing the layers to that eyes and mouth and other features line up in a way that adds up to a new coherent portrait.
Since I am using a computer program I am looking for a way to create the same level of coherence. Up until now I had to prepare the source images by cropping them to faces like in the Lara series. I was searching for a way to make this process automated as well and then I found that you can add facial recognition modules to the processing programming environment. I could use that to have the program crop the images for me. I collected images of Bella Hadid and had the program crop them and combine them using the original technique that I invented for the Kylie portrait.
The resulting images to me are very convincing portraits of Bella Hadid. They have an unexpected variance in the series. But the consistency in cropping creates a very consistent composition. This motivated me to start creating more Photoshop based portraits as well, in which I try to create a bit more expressive and extreme portraits.
I decided to keep using the facial recognition cropping of the images, as I feel it really ads to the conceptual starting point of these portraits: to be as hands-off as possible and use as much pre existing images and choices, to try and make new and iconic portraits.







