Still Life no. 09 (2019). Artists learn from past artists that set the groundwork of artistic ideas before them. In this series of still-life paintings, I am learning from 20th century painters by automating the process. Recently, I worked with a painting mentor who suggested I look at old masters artwork. By spending time with the masters and copying their paintings, she suggested I would develop painting techniques that can then be applied to my own art practice. I had an idea of how to automate this common artistic practice with classifiers and GANs to create work that is not direct copies.
I wrote Python code that generate still life compositions by randomly placing colorful background shapes, vases, fruits and flowers. For example flowers must be placed directly above a vase. These generated compositions became my dataset to be processed with a cycleGAN to make them painterly looking. My painting dataset included: Cezanne, Matisse and Canadian artist Joseph Plaskett. Now with thousands of generated digital painting images, I used a TensorFlow Classifier, trained on the painting dataset to sort or “curate” the unique images. With the images that are rated the most confident to be closes to the painting dataset paintings, I sit down and physically paint them. As I paint more images, I retrain the classifier and cycleGAN to use my paintings rather than other artists and in turn developing my own style.
Paintings created in this project: https://joannehastie.com/ml-still-life/
More about this project: https://joannehastie.com/project/learning-from-matisse/
