Can you make a jigsaw puzzle with two different solutions? Or an image that changes appearance when flipped? We can do that, and a lot more, by using diffusion models to generate optical illusions! Continue reading for more illusions and method details 🧵

RT @ZhaoyingPan: Our workshop at #ICLR2025 is now open to submissions until 02/03 🥳 check our website if you are interested:

RT @davidrmcall: Decentralized Diffusion Models power stronger models trained on more accessible infrastructure. DDMs mitigate the networ…

Hey all, I'll be answering questions about our "Motion Prompting" paper on alphaXiv (@askalphaxiv) (it's like arXiv, but adds a discussion section, and I think is quite well built!):

url: repo: hopefully you find this useful! let me know if there are problems, if somebody has done this before, or if you have any other comments!

I had a lot of fun helping put this problem set together -- if you're teaching diffusion models + computer vision, consider using this homework for your course! (links at end of @ryan_tabrizi's thread!)

Oh, and Tatiana's twitter profile is @ZePoLiTaT! :)

I'll be presenting "Images that Sound" today at #NeurIPS2024! East Exhibit Hall A-C #2710. Come say hi to me and @andrewhowens :) (@CzyangChen sadly could not make it, but will be there in spirit :') )

RT @nico_dufour: 🌍 Guessing where an image was taken is a hard, and often ambiguous problem. Introducing diffusion-based geolocation—we pre…

@sammtmd Thank you! Sadly, there are no plans to release it :')

RT @xunhuang1995: 🚀 Introducing CausVid: Instant video generation that plays the moment you hit "Generate", while maintaining state-of-the-…

RT @PardoAlejo: Ever noticed how one scene seamlessly transitions into another in films? That’s a match-cut—a subtle yet powerful cinematic…

@natanielruizg Thanks Nataniel!!

@sherwinbahmani Thank you Sherwin!!

RT @chang_trenton: (1/) I'll be going to #NeurIPS2024 next week, where I'll be presenting exciting new work on detecting gaming using causa…

@_amirbar @OfirPress @JunhwaHur @forrestercole @CarlDoersch @yusufaytar @MikiRubinstein @jesu9 @oliver_wang2 @andrewhowens @DeqingSun Thank you!!

@adityakusupati Thank you so much Aditya!!

@distributionat wow thank you very much thomas sir

@fogggysuki Thanks! It sadly is not real time :') but we're hoping these results will drive more interest in getting something like this real time!