60-212: Interactivity and Computation for Creative Practice
CMU School of Art / IDeATe, Fall 2020 • Prof. Golan Levin
Artbreeder was great to create rich imagery with endless possibilities. After some experimentations, I liked images that had realistic lighting with unconventional forms/textures. I felt a bit guilty while making the images though, as I felt like I had too much power creating impressive surrealist images only at the touch of a button.
The model was not given an “environment” class which just composes images it is not supposed to recognize as other classes. As such, it recognizes empty space as car keys even if there are no car keys present.
I am pretty impressed by how quickly this model is created with reasonable results. Reminds me of the quick training models that azure custom vision offers.
It’s shocking how the network switches from personal narrative to informative third person based on the prompt. I want to experiment more by feeding prompts toeing the line in-between to see any biases in the training set to favor 1 style of writing versus the other.
Generated text:
Before leaving to college my mom told me shocking news. My grandmother who is over 90, and doesn’t see eye to eye with any of us, wanted to live with us to help take care of my grandfather. I couldn’t believe it! There is so much emotional baggage that my mom is holding on to that she doesn’t realize she’s causing. The other thing that has gotten me stressed out is her shopping spree! While I was in town she went out shopping and bought clothing for my grandmother that wasn’t in her wardrobe. While I was home she had my grandfather take her
another example:
He was impeached. [97] [98] He died in London in 1866, unmarried. He was second son of Charles Gavan Duffy, and brother of Charles Gavan Duffy, 18th Baron Ardee. “The Mannings of San Francisco” (published 1898) includes an account of this group of characters. The words used here are taken from the 1884 memoirs of Mary and Tom Manning, which were originally published in the “San Francisco News”.
My guess on how these models were trained is that an edge detection algorithm was run on cat images and that input was fed into the model and the output was expected to be the original cat image. As such, this model won’t produce realistic cat features with hand drawings alluding to them. In the image below the triangle eyes are just converted to fur instead of eyes. The same lack of recognizing hand drawn features persists. Still a fun tool though.
I played with this tool. Crossing the midway when determining cross breed proportions causes more change than any other shift in position. The breeder is very good at creating believable images, to which it sacrifices user control on the combination process. The more dissimilar the more the product is unpredictable.
=
This is so natural…!
=
50 : 50 mix looked too realistic to be interesting so I made it 70 : 30.
Plumbing the depths of latent space
Stacked interpolation cycle – StyleGAN × floor plans pic.twitter.com/hKhYtQJ9Po— Erik Swahn (@erikswahn) May 17, 2019
This work by Erik Swahn depicts many floor plans stacked on top of one another. It was made using a StyleGAN (generative adversarial network). I chose this work because it is really satisfying to look at cross sections of things you wouldn’t normally see. I think that this work is especially imaginative because it paints a picture of a building that does not exist, so legitimately trying to imagine what this creation would look like from how we normally interact with buildings is a fun challenge. I think this technique of interpolated layers has a similar aesthetic to Robert Hodgin’s Meander that we saw earlier this year. I also just really like how this looks because of the way it is rendered; normally floor plans are so boring
https://teachablemachine.withgoogle.com/models/6U2T8odGG/
I definitely want to experiment with all the other types of templates they offer. It was really easy and intuitive to use and I wonder how it would react if I draw the plushies.
Once upon a time. “Logan?” I squeeze his hand, and he squeezes back, lifting his head to stare at me. “What is it?” His eyes are blank, unreadable. He shakes his head, giving me that playful grin of his that I love. “It’s nothing, princess. I was just, um, thinking of an old girlfriend. No big deal.” He gives me a smirk, and the jerkiness of my heart vanishes. “It’s nothing to worry about. It was a long time ago, and you know, we all have to move on.” “No!” I cry. “You can’t say that.” “But it’s…
Donald Trump is in a lot of trouble after losing the first presidential debate. At the same time, Trump is on the rebound in the polls, and he is even leading in some. But the bottom line is that Trump has to turn things around, and right now, he’s not doing a good job. In fact, Trump did the exact opposite during Monday’s debate. Instead of trying to hold his own, Trump followed Hillary Clinton’s lead, and it blew up in his face. On several occasions, Trump made it clear that he simply didn’t know what he was
The dress was gold and white She stood in the middle of the floor Her body locked in place. She waited for him to do what he always did. “Well, well,” he said and his eyes got darker and darker, “you were a sight to behold. After what you did to those pants,” he said and pushed his finger into his mouth, “I thought they’d be impossible to fit.” “Sorry,” I said. I kept my head down, not wanting to look up and be caught with his dirty, thirsty eyes on me. “Yeah, that’s what I thought,” he said, as he moved down the
I think Artbreeder is a super cool tool. I found myself really getting immersed in the infinite possibilities. The animation feature was really nice as well. I think that it is certainly a fun creative tool, but I also feel like at some point, the algorithms do all of the work. Combined with the sheer abundance of work, it is difficult to see the creativity in the output.