[Written by ChatGPT. Main image: “flowers in the pacific northwest, realism” (SD 2.1)]
Spring has finally arrived here at Neural Imaginarium, unfurling its vibrant hues after a particularly long winter. It’s a universally appreciated respite, a breath of fresh, blossom-scented air that fills us with renewed energy and excitement. Inspired by the long-awaited arrival of the season, we embarked on an exploration of how DreamStudio, our resident AI artist, handles the task of reimagining famous floral artworks.
Our journey started with a minor hiccup. ChatGPT, our trusty AI writer, offered up a piece titled “Bouquet of Sunflowers” by Henriette Wyeth. However, we were unable to find any such work by the artist. Despite this, our AI artist did not disappoint and delivered on the sunflower theme as described in the prompt.
[“Large sunflowers and other blooms in a ceramic pitcher, rich colors, realism, Henriette Wyeth.” Clockwise from left: SD 1.5, 2.1, XL]
As we progressed through the art pieces, the results were quite fascinating. From Van Gogh’s sunflowers rendered in a fresh and interesting way to our own AI-driven interpretation of Monet’s “Poppies”, with an unexpectedly more focused depiction of human figures, we found the results to be intriguingly different.
[“Vase with sunflowers, vibrant yellows and greens, post-impressionism, Vincent van Gogh,” same pattern as above]
[“Field of red poppies, mother and girl, impressionism, Claude Monet,” same pattern as above]
One observation that stood out was the slightly digital aesthetic of the results from the XL model of Stable Diffusion. Detail lines were noticeably clear, even in the impressionist works. While this distinctive style isn’t inherently unpleasant, it certainly deviates from the more traditional, softer approach of Impressionism.
Overall, the experiment was a colorful exploration of the blossoming AI artistry. And we want to make it a bit more fun for you, dear reader. We’ve compiled some of the generated images for a guessing game! With some lesser-known works in the mix, we’re rating the difficulty as “medium”. For those of you well-versed in the art world, we suspect you might find it a tad easier.
[Each row represents a single prompt. Left to right: SD 1.5, 2.1, XL]
Categories: Image
Leave a Reply