The idea of 'machine unlearning' is getting attention lately. Been thinking a lot about it recently and decided to write a long post: ai.stanford.edu/~kzliu/blog/un… 📰

Unlearning is no longer just about privacy and right-to-be-forgotten since foundation models. I hope to give a gentle…

🪟 x 🖼️

Adobe published GS-LRM!

The method can generate high-quality 3D Gaussian primitives from just 2-4 posed sparse images in 0.23 seconds on a single GPU 🔥

6 interactive examples ⬇️

Washed Out 'The Hardest Part'

I leaned into the hallucinations, the strange details, the dream-like logic of movement, the distorted mirror of memories, the surreal qualities unique to Sora / AI that differentiate it from reality. Embrace the strange.

We are entering a new era of production in the music industry.

MGMT’s “Mother Nature” music video was created using Move AI for motion capture.

This is motion capture that generates 3D motion data from 2D video. No suits required.

Made this last night on unreal engine 5 🙌🏻

This entire 3D game environment was created from one 2D skybox. How do you like that parallax?? See original skybox below. 👇

#madewithunity #skyboxAI

New nodes for weight scheduling. IPAdapter, text prompt and controlnet all sync'ed together. Have fun! youtu.be/jc65n-viEEU

painting flowers with ai, irl

how to below

Sorry 😬I know Its really late notice PLUG !
It'd be great to share some of the stuff I've been up to.

'🎨🗣️ Excited for a digital rendezvous and insightful conversations @24hopestreet this Friday, May 3rd, starting at 7 pm! 🌟 See you there! '
#DigitalArt #Conversations #BYOB

This is hand and finger motion capture powered by AI.

3D motion data is generated from 2D video. No suits or gloves required.

This example was captured using Move Multi-Cam with 8 GoPro cameras shooting at 4K and 120 FPS.

3D Gaussian Splats are cool, but they're static (Part 26).

PhysDreamer is a physics-based approach that enables you to poke, push, pull and throw objects in a virtual 3D environment and they will react in a physically plausible manner.

5 wild examples + links ⬇️

A little sneak peek at my upcoming tutorial!!!

Learning to control and manipulate lighting in your GenAI images using #Substance3D Stager and #Adobe Firefly's Structure Reference!!

This tutorial is the next step in my series on utilizing 3D workflows to gain some control over…

This is what Siri should do but open source did it first.

Testing disco diffusion 5.2 Warp. too good! Thanks to Alexander S. DD offers so many possibilities! Want to make a short film with this technology.

Another Wonder Dynamics animation. I cleaned it up in Blender and animated the sword. I originally wanted to replace the background but since there was a lot of camera movement, the characters feet weren’t at a consistent height. There’s probably a way to fix that, just haven’t…

What if you could use AI as a Render Engine? 🤯

Check out this video where I replace Captain America with someone you may know 😉 (🔊 on)

If you enjoy this content please consider liking and sharing - it really helps me grow!

As always, details below 👇

📢 On 1 May we'll be opening applications for round 2 of R&D grants available for university-led projects that explore novel use cases & applications of VP & XR tech. Take a look at the projects supported through round 1 of funding & stay tuned for updates bit.ly/3w2ZWst

RefFusion is a 3D scene inpainting method that can insert objects into a scene and outpaint it to complete it.

reffusion.github.io

introducing aspect ratios in Realtime.

now you can generate horizontal and vertical images.