MDM: Human Motion Diffusion Model

Re:MDM: Human Motion Diffusion Model by obiefernandez on Thursday September 29, 2022 @09:04PM (#33029533)

Just me or is something new like this popping up every day now?

[ Parent | on HN ]

Re:MDM: Human Motion Diffusion Model by ipsum2 on Thursday September 29, 2022 @11:35PM (#33030595)

It's because there was a major ML conference submission deadline yesterday, so now things are being announced.

[ Parent | on HN ]

Re:MDM: Human Motion Diffusion Model by lajamerr on Friday September 30, 2022 @02:36AM (#33031685)

https://iclr.cc/

The conference for reference

[ Parent | on HN ]

Re:MDM: Human Motion Diffusion Model by RosanaAnaDana on Thursday September 29, 2022 @11:12PM (#33030441)

Its been about every 5-6 hours today.

[ Parent | on HN ]

Re:MDM: Human Motion Diffusion Model by uni_rule on Thursday September 29, 2022 @10:04PM (#33029998)

The pure volume of projects spurred on by the release of Stable Diffusion source code is staggering

[ Parent | on HN ]

Re:MDM: Human Motion Diffusion Model by ShamelessC on Thursday September 29, 2022 @10:22PM (#33030131)

The original diffusion code for these projects comes from lots of research. In terms of code- OpenAI released guided-diffusion over ImageNet classes awhile ago, then GLIDE, their DALLE2 predecessor. CompVis incorporated that into their Latent Diffusion variant about half a year ago. A few weeks before Stable Diffusion released they also published the similar “retrieval augmented diffusion” codebase.

And _then_ stable diffusion came out. The effectiveness of diffusion models for text to image is an idea that has been floating around a little longer than you might think.

[ Parent | on HN ]

Re:MDM: Human Motion Diffusion Model by Vt71fcAqt7 on Friday September 30, 2022 @10:23AM (#33035249)

I think I agree here. Stable diffusion is "just" an optimized version of DALLE2, which itself directly builds on the previous literature. Also, the 3d model generator posted to HN yesterday builds on NeRF. And they are all using CLIP I believe. Stable diffusion gave us speed and knowlege about how much data of what kind and quality needs to be in the training data to still generate good results.

[ Parent | on HN ]

Re:MDM: Human Motion Diffusion Model by ShamelessC on Friday September 30, 2022 @02:22PM (#33038603)

It's certainly not my intent to undermine the efforts of Robin Rombach, Andreas Blattman, Katherine Crowson (and many others).

Katherine's work on clip-guided-diffusion over the `guided-diffusion` ImageNet checkpoints was effectively the first time the public got to see what text-to-image via diffusion instead of purely transformer-based solutions (like in DALLE1/dalle-mini) would look like. And it happened well before GLIDE was published (and gets a mention/citation).

The CompVis team (Blattman, Rombach, etc.) has been able to not just compete, but surpass (in some ways - it's nuanced) the work of the big American research labs (OpenAI in particular) with solid novel research. Their research on `VQGAN` outperformed the Autoencoder from the DALLE-1 paper, and they've been competing directly in the vision space ever since.

Incredibly talented people.

[ Parent | on HN ]

Re:MDM: Human Motion Diffusion Model by antegamisou on Thursday September 29, 2022 @10:36PM (#33030204)

Yet the SNR is getting abysmal day by day.

[ Parent | on HN ]