CREATING ART
Creativity is a long game
Your past work is always something to build upon
“You have just produced your masterpiece.” — My inner voice whispers this sentiment every time I pour myself into creating a work from a blank canvas, real or virtual.
This inner voice, inclined to delusions, is usually just trying to keep my spirits up.
Creating can be hard. Well, creating stuff isn’t hard so much as it is like hanging your underwear out to dry; people can be critical about your intimate expression.
Still, I create in the face of judgement. I know that in making (and sharing) something, I am generating the seeds for future work.
Ideas live in both the best and worst of our creations.
To give an example of what I mean, I thought I would share the steps in the journey to create a painting that sparked my transformation from scientist to visual artist.
The beginning: doodles
I worked as a laboratory scientist most of my career, so I created art infrequently. My free artistic expression mainly came out in a habit of doodling in the margins of my notebooks during lab meetings.
I had a couple of rough years where each week began with a contentious lab meeting on Monday morning. My doodling became an important tactic to hold my tongue and stay focused. I did a lot of doodling.
When I decided to quit my biotech career at midlife, I wanted to memorialize the impetus for leaving. I decided to do something with the doodles. I assembled a collection of the doodles in a collage I called Monday Morning Meetings.
My inner voice told me it was a masterpiece.
Well, the collage is not the best, but layering all that doodle emotion together felt amazing. This piece was a seed that made me want to do more.
Step two: self-portrait
My doodle collage led me to think about drawing again. I have always liked to draw, but my focus was more concentrated on science than art since my college days. The doodle collage inspired me to sign up for an eight-week life drawing class.
I was a bit out of my league in the class. Most of the students were sharpening their skills for industry jobs at Disney or in video game design.
While taking the class, one of the homework assignments was to sketch a self-portrait using a mirror instead of a photo. Drawing from real life is challenging because one must capture perspective and dimensions quickly before the subject moves.
I had a bit of an extra challenge doing my self-portrait. I needed reading glasses to draw on my paper, but I had to take the glasses off to see myself in the mirror. The resulting portrait echoes my visual dilemma, which can be seen in the wonky rendering of my eyes.
My inner voice told me it was a masterpiece.
Alas, the sketch is very naive without shading or depth, but I was proud that I completed the assignment and didn’t drop out of the class.
Step three: style transfer with GANs
The drawing class had me thinking about my limitations. I enjoyed creating, but my pure drawing skills were years behind the plethora of young-eyed artists seeking their destinies.
I started reading a lot of articles about artificial intelligence (AI)-generated art. The scientist in me was intrigued. I wanted to understand how I could use the power of AI to enhance my own skills.
I learned about generative adversarial networks (GANs) and how GANs were being used to apply various artistic styles to photos. I even downloaded some notebooks from GitHub to play around with Deep Dream.
At this step in my journey, I found DeepArt.io, a free, neural network style transfer platform co-founded by the inventor of the neural style transfer algorithm, Leon Gatys.
I like that the DeepArt.io desktop app lets you upload your own style to transfer. I decided to transfer the style of Monday Morning Meetings onto Self-portrait.
Both my inner and outer voices thought that what the algorithm generated was a masterpiece.
The generated image kept the integrity of my self-portrait but etched doodle lines and colors onto my face that evoked all the emotions I felt in the eight months since I left my job.
One thing that bothered me was that I didn’t create the output myself. I directed the output, but the computer did the rendering.
Step four: the painting
The journey I am sharing predates many of the recent developments in AI-generated art. With the advent of Open AI’s DALL-E 2 and derivatives, anyone can generate beautiful renderings of whatever they want in seconds.
I find AI-generated art revolutionary. What I see missing in prompt-based AI-generated art is the tactile component of creation.
I decided to use the DeepArt.io output of my Monday Morning Meetings Self-portrait hybrid as a reference for my own mixed media painting.
I still remember drawing the lines, mixing the colors, and layering the paint. The computer doesn’t give me this feeling.
The resulting painting, Time is 8 months, is the culmination of all the previous inputs, a collaboration between the scientist, the computer, and the artist.
The cool thing about the outcome is that the result gave me courage to enter Time is 8 months into a juried art show at a gallery in Laguna Beach, CA.
The painting was accepted into a group show. I was so excited.
Continuing steps
The total time from step one through step four took about two months. If you count the doodle-making, the whole process really took more than 10 years.
What I relearned during this journey is that creation not only pulls from multiple experiences, but it is also builds upon previous efforts.
If you are doubting your voice as an artist, keep creating. Imagine then reimagine until you find the next step that moves you forward.
This starting creative journey provided me the validation to continue along a path committed to telling stories with visual arts.
Julia Fletcher is founder of JEFS Storytelling Arts, a graphic design studio, where she uses her unique research skills and artistic talents to create custom visual stories that help clients’ increase engagement and promote the education of their audience.