This project was made by Fabian Stelzer to test out the abilities of various text-to-image AIs.
Tech Artist and AI enthusiast Fabian Stelzer has unveiled the results of a cool experiment with the Midjourney text-to-image AI. This time, the artist has managed to set up a simple character animation using the aforementioned network to generate the old man portrait, Avatarify, a neural network-powered app that can put pre-recorded expressions on a photo, to give the man facial animations, and Synthesia, to set up the character's voice. In the end, the artist created a fully-generated animation that was made almost entirely by different AIs.
This project is a part of Fabian's absolutely enormous "DALL-E 2 vs Midjourney vs StableDiffusion" Twitter thread, a lengthy series of posts dedicated to comparing and contrasting the AIs and showing how they interpret the same prompts. If you are a fan of generative art, we highly recommend checking out the full thread by visiting the author's Twitter page.
Also, don't forget to join our Reddit page and our Telegram channel, follow us on Instagram and Twitter, where we are sharing breakdowns, the latest news, awesome artworks, and more.