Time to share our #SIGGRAPH2024 paper "N-Dimensional Gaussians for Fitting of High Dimensional Functions"! We show that N-Dimensional Gaussian mixtures can be optimized to handle high dimensional inputs like MLPs with a fraction of the training times. sdiolatz.info/ndg-fitting/

Introducing Cambrian-1, a fully open project from our group at NYU. The world doesn't need another MLLM to rival GPT-4V. Cambrian is unique as a vision-centric exploration & here's why I think it's time to shift focus from scaling LLMs to enhancing visual representations.🧵[1/n]

Accepted to #ECCV2024, version 2 of our past 'Accidental Light Stage' work (grail.cs.washington.edu/projects/Light…). We can perform real-time temporally consistent relighting by using monitor relighting in any condition (*as long as the monitor light has some relative strength).

Thank you Jason for sharing! This is an amazing project!! Idle GPU to browse YouTube for you, now! lol

New project delivered😍 PlanPoint architectural visualization using #threejs + #r3f is live. An entire house in less than 15 MB. Sharing some tips for good performance Link: planpoint-house.vercel.app Let me know how it’s performing in the comments #threejsJourney #creativecoding

Virtually hugging Einstein! My intellectual hero!

Shape of Motion: 4D Reconstruction from a Single Video arxiv.org/abs/2407.13764 Project: shape-of-motion.github.io Code: github.com/vye16/shape-of… Great #3DGS-based reconstruction from monocular videos! Method ⬇️ 1 I 2

The mix of speed + smarts on display has me shook. Rare AI demo to have this effect 🤯 This is what iteration at the speed of thought looks like. Like, how can this tech not absolutely transform knowledge work?

Depth-of-Field representation in wireframe space

Almost time for another year of #SIGGRAPH2024 !!! youtu.be/tjYVcOJONdI?si…

This is wild. You can now run Llama 3.1 405B on 2 MacBooks at home in AI cluster 🤯

Whoa looks like there's gonna be a Starline telepresence demo at SIGGRAPH next week s2024.conference-program.org/presentation/?… I haven't demoed Starline in a while but all of my early demos were extremely impressive, comparable to the feeling of trying VR for the first time.

What if a real-world object has a context menu? Check out #AOI - Augmented Object Intelligence that brings #AI to everyday objects in #AR: fully opensourced at xr-objects.github.io Thank you Dr. Doga Dogan for your amazing work @google! UIST 2024 #UIST2024 #HCI #LLM #XR #VR

OmniRe: Omni Urban Scene Reconstruction About the Project: "We build a dynamic Gaussian scene graph leveraging the collective strengths of different representations: • Vehicles are modeled as static Gaussians, transformed using rigid body transformations to simulate their

In the future remote workers will be able to do some serious damage. Here’s me having a bad day while piloting the 1x robot 😂

Loopy: New Audio-to-Video Lipsyncing Model Looks Insane It generates lifelike facial expressions and movements from audio alone. It captures subtle details like sighs, expressive eyebrows, and natural head gestures, making your videos incredibly realistic. Sample videos 1/5

A few prompt engineering tips I picked up during my time at OpenAI

AI tools have no creative control; they're like slot machines. Yeah buddy, sure. This is ReshotAI. In the coming months, we will see many more tools like this. Future of AI is bright ✌️

It's dying again... Shall we use md5 next time as ACM SIGGRAPH? 😂