iiinova — where your digital self keeps evolving

Ben Poole

@benmpoole.bsky.social

1.9K followers 231 following 6 posts

research scientist at google deepmind. phd in neural nonsense from stanford. poolio.github.io

Posts Replies Media

Reposted by Ben Poole

Arthur @arthurperpixel.bsky.social · 30/01/2025

I’ve released a new version of my 3D reconstruction tool, Brush 🖌️ It's a big step forward - the quality & speed now match gsplat, and there’s a lot of other new features! See the release notes github.com/ArthurBrusse... Some of the new features:

A autumnal stump, covered in mushrooms. This is a still from the interactive 3D reconstruction!

2278

Reposted by Ben Poole

Hernan Moraldo @hmoraldo.bsky.social · 16/12/2024

Physics with Veo! #veo2 Prompt: One ball is in the floor. Another ball comes rolling.

371

Ben Poole @benmpoole.bsky.social · 12/12/2024

Come learn about CAT3D today at #NeurIPS2024! talk: 3:30pm West Exhibition Hall C, B3 poster: 4:30pm East Exhibit Hall #1610 remote: cat3d.github.io we've got stickers 😸

1110

Reposted by Ben Poole

Zhenjun Zhao @ericzzj.bsky.social · 11/12/2024

SimVS: Simulating World Inconsistencies for Robust View Synthesis Alex Trevithick, Roni Paiss, Philipp Henzler, Dor Verbin, Rundi Wu, Hadi Alzayer, @ruiqigao.bsky.social, @benmpoole.bsky.social, @jonbarron.bsky.social, @holynski.bsky.social, Ravi Ramamoorthi, Pratul P. Srinivasan

1122

Ben Poole @benmpoole.bsky.social · 04/12/2024

Information in observed images that predicts what should be in unobserved regions should be captured by these models (e.g. x.com/DotCSV/statu...). They can't create new information, but there are often more statistical hints than we as humans perceive (like the paper you linked).

x.com

000

Reposted by Ben Poole

ruiqigao.bsky.social @ruiqigao.bsky.social · 02/12/2024

A common question nowadays: Which is better, diffusion or flow matching? 🤔 Our answer: They’re two sides of the same coin. We wrote a blog post to show how diffusion models and Gaussian flow matching are equivalent. That’s great: It means you can use them interchangeably.

625459

Ben Poole @benmpoole.bsky.social · 02/12/2024

eh, perception as probabilistic inference is a well established theory of the mind

110

Ben Poole @benmpoole.bsky.social · 01/12/2024

This is the essence of visual perception. We don't see pixels, we experience the world behind an image. Our aim is to build AI systems that can achieve this same level of spatial intelligence. Inferring the whole from fragments requires statistical priors for both humans and machines.

120

Ben Poole @benmpoole.bsky.social · 01/12/2024

unlike most work on image/video editing that is not guaranteed to "follow rules" of 3D consistency, our work builds an explicit virtual world that you can move a virtual camera through. we use a statistical learning-based system, but we distill it into a 3D model that follows at least some 3D rules

130

Reposted by Ben Poole

Aleksander Hołyński @holynski.bsky.social · 28/11/2024

Check out CAT4D: our new paper that turns (text, sparse images, videos) => (dynamic 3D scenes)! I can't get over how cool the interactive demo is. Try it out for yourself on the project page: cat-4d.github.io

16314

Ben Poole @benmpoole.bsky.social · 28/11/2024

Stop watching videos, start interacting with worlds. Stoked to share CAT4D, our new method for turning videos into dynamic 3D scenes that you can move through in real-time! cat-4d.github.io arxiv.org/abs/2411.18613

29014