🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

#CVPR2023 is mostly posters this year, which is good! Poster sessions can be great. But a lot of people are not so great at presenting their poster. Here's some of the advice I give grad students when preparing for poster sessions.

Really cool to see an open-sourced version of Neuralangelo 🏛 already out just a few weeks after our release! Congratulations @AutoVisionGroup and the SDF studio team!!

Epipolar geometry FTW! #3DV2024

Will we be kicked out of the room if we use epipolar constraints in diffusion models? 😂 Check out our CVPR paper on pose-guided diffusion models for novel view synthesis! poseguided-diffusion.github.io

Happy to share our #ICCV2023 paper! Let me know, if you have any questions or want to discuss any details :) Thanks for the nice collaboration! Ang Cao Andrew Owens Justin Johnson Matthias Niessner

Super excited to share our work SIGGRAPH Asia ➡️ Hong Kong: 3D Human Digitization from a *Single Image* 💡 Core idea: Synthesize consistent and detailed appearances using a 3D shape-guided diffusion model!

"Sora Generates Videos with Stunning Geometrical Consistency", neat result. Seems obviously true from just looking at the Sora results, but having metrics is important. sora-geometrical-consistency.github.io

ViewDiff 3D-Consistent Image Generation with Text-to-Image Models 3D asset generation is getting massive amounts of attention, inspired by the recent success of text-guided 2D content creation. Existing text-to-3D methods use pretrained text-to-image diffusion models in an

Check out our #CVPR2024 work on converting a 2D text-to-image diffusion models into a text-to-3D generator!

Congrats to AI at Meta on Llama 3 release!! 🎉 ai.meta.com/blog/meta-llam… Notes: Releasing 8B and 70B (both base and finetuned) models, strong-performing in their model class (but we'll see when the rankings come in @ LMSYS Org :)) 400B is still training, but already encroaching

Better datasets are a major driver of advances in AI. Our new DiffuRank project provides a dataset of 1M (text, 3D object) pairs by captioning Objaverse objects; the biggest open dataset of its kind. Try it out for scaling up your favorite methods in generative 3D!

Amazing review verbatim - I'm speechless. The computer vision community should be ashamed that this is the state of our peer review system "Strengths: A good analysis is performed and some conclusion are been drawn. "Weaknesses: These more of an incremenatal work no detailed

We are looking for 2025 summer interns in Reality Labs at Meta! Please apply using the below link and reach out to me if you wish to work with us on topics related to 3D scene reconstruction, neural rendering, Gaussian Splatting, or real-time graphics. metacareers.com/jobs/117689680…

Our NeurIPS paper is published on arXiv. In this paper, we propose a new optimizer ADOPT, which converges better than Adam in both theory and practice. You can use ADOPT by just replacing one line in your code. arxiv.org/abs/2411.02853

Introducing Generative Omnimatte: A method for decomposing a video into complete layers, including objects and their associated effects (e.g., shadows, reflections). It enables many cool applications, such as video stylization, compositions, moment retiming, and object removal.

3D Gaussian Splatting is awesome! BUT, the 3D Gaussian primitives often have difficulty reconstructing fine appearance details. 😩 Meet Textured Gaussians! ✋ Our RGBA Textured Gaussians enhance 3D scene appearance modeling. Small change, BIG improvement!

Today we're sharing our first research update World Labs -- a generative model of 3D worlds! I'm super proud of what the team has achieved so far, and can't wait to see what comes next. Lifting GenAI to 3D will change the way we make media, from movies to games and more!

Introducing MegaSaM! 🎥 Accurate, fast, & robust structure + camera estimation from casual monocular videos of dynamic scenes! MegaSaM outputs camera parameters and consistent video depth, scaling to long videos with unconstrained camera paths and complex scene dynamics!