Stable Diffusion: 8 New Amazing Results!

📝 The paper "Diffusion Self-Guidance for Controllable Image Generation" is available here:

Try it out!

Sources:

My latest paper on simulations that look almost like reality is available for free here:

Or this is the orig. Nature Physics link with clickable citations:

Thumbnail background image: sutrik

🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible:
Aleksandr Mashrabov, Alex Balfanz, Alex Haro, Andrew Melnychuk, Benji Rabhan, Bret Brizzee, Bryan Learn, B Shang, Christian Ahlin, Eric Martel, Geronimo Moralez, Gordon Child, Jace O'Brien, Jack Lukic, John Le, Kenneth Davis, Klaus Busse, Kyle Davis, Lorin Atzberger, Lukas Biewald, Martin, Matthew Valle, Michael Albrecht, Michael Tedder, Nikhil Velpanur, Owen Campbell-Moore, Owen Skarpness, Rajarshi Nigam, Ramsey Elbasheer, Richard Sundvall, Steef, Taras Bobrovytsky, Ted Johnson, Thomas Krcmar, Timothy Sum Hon Mun, Torsten Reil, Tybie Fitzhugh, Ueli Gallizzi.

Károly Zsolnai-Fehér's links:

[sapphyrus]

I love how I can use Stable Diffusion on my own computer, very happy to see it developing like other online-only options!

[eleonora]

The QR-code art is really a cool idea. I guess this could become a trend someday

[randomdude]

The QR codes actually work I cant hold my papers anymore 🤯

[mr.mirror]

can't wait to grip these papers like a pull up bar 😭😭

[QuantumCrafts]

Stable Diffusion: Where AI meets artistic wizardry, turning childhood doodles into masterpieces and even making QR codes a fashionable form of art!

[jonmichaelgalindo]

I love open source AI!!! It makes me so hopeful for the future. 🥰 (I try to contribute, but every time I work on a project, someone beats me to the punch. I'll still do what I can though!)

[Chris.Davies]

I am a simple man: Karoly uploads a video. I watch the video. I marvel at the video. I upvote the video.
Simple!
Edit: Now I'm worried: AI-generated hotness can be a little disturbing! No more uncanny valley!

[dissonanceparadiddle]

This is so impressive!!! I'm excited to give it a try.

[rogueyun]

I love how that as he crushed the can it tried to turn the can into an anime girl... That was insane.

[TheMaxxxel]

This childhood drawing is better than everything I've ever drawn

[Djellowman]

Insane! And the QR-code art is very cool

[RandomGgames]

You know what I haven't seen much of yet which surprises me? Taking a video walking around an object and directly extracting a 3d model of it, such as an OBJ or STL. That technology would be amazing. Does that exist/are there any tutorials for it out there? I've been using photogrammetry a lot for personal projects however it's not always the best.

[randomdude]

Stable diffusion just keeps stably diffusing

[Will_of_Iron]

Please make a video on the revolutionary liquid neural network! I'm surprised that you haven't made a single video on that topic! It's a huge request from me!!😊🙏

[bambabam]

As always thanks for a great breakdown of current state of the art technology. Do you have a link for the google paper you mentioned at the end?
Instructing Text-to-Image AIs with natural language would be a huge step, if they actually manage to keep coherence, especially in characters

[gressvikgutten]

Exiting! But i cannot help but thinking that catfishing just took a huge leap forwards 😬😂

[OubaydaKhayata]

1:57 "Hold on to your phones Fellow Scholars!" 😄

[mitch]

When the computer I used for almost ten years finally died, I tailored its replacement to run Stable Diffusion. I love it. I love the state of technology today.

[sebastianhernandez]

Could you please provide the link to the account that works with TextVideo? It's not in the description.

[DJVARAO]

SD advanced thanks to the open source work of many scholar enthusiasts.