Excited to share our #NeurIPS2022 @NVIDIAAI work GET3D, a generative model that directly produces explicit textured 3D meshes with complex topology, rich geometric details, and high fidelity textures. #3D Project page:

RT @zianwang97: 🚀 Introducing DiffusionRenderer, a neural rendering engine powered by video diffusion models. 🎥 Estimates high-quality geo…

RT @xuanchi13: 📢🚗✨ Excited to announce InfiniCube, our scalable generative model for dynamic 3D driving scene generation with high fidelity…

RT @TianchangS: Generating nice meshes in AI pipelines is hard. Our #SIGGRAPHAsia2024 paper proposes a new representation which guarantees…

One thinking after reading this: Will the research to many problems in this world become: (a) how can we convert the problem to a computational structure for AI, and (b) how can we prepare the data for training AI?

Congratulations! I think the most exciting part from this award (and also from Alphafold) is providing acknowledgment, confidence and passion to broader AI researchers, the goal of research is far beyond publishing some papers!

RT @zianwang97: Can 2D diffusion models help 3D editing? We use diffusion models as data priors for inverse rendering, inspecting the edi…

@Summer912_ I had the same feeling of stress and anxiety last year. The tips I used: whenever have such a feeling, read our statements to gain confidence about ourselves. It helps to practice "fake it until you make it" eg constantly tell myself "I'm the best", and ask friends to tell me.

If you are at SIGGRAPH, check out NVIDIA kaolin hands-on lab! It will cover many features for 3D deep learning in PyTorch, including diff render, and phys sim of many representations, including GS and meshes, in just a few lines of code, and in Jupyter notebook! Thur, 2:15pm!

RT @_shumash: Join hands-on lab at #SIGGRAPH to learn how to simulate points, 3D gaussian splats, meshes in just a few lines of code and ri…

RT @xuanchi13: Our code for XCube is released! Check it out and give it a try! Also come to our poster session #C…

@sellan_s @Columbia Congratulations!

RT @jonLorraine9: New #NVIDIA #GTC24 paper 🎊 We generate high-quality 3D assets in only 400ms from text by combining (a) amortized optimiz…

@Michael_J_Black @SIGGRAPHAsia @Marilyn59846278 @KeenonWerling Congratulations!!!!

@MattNiessner @SIGGRAPHAsia @MZollhoefer @izadi_shahram @mcstammi Congratulations!!

Proud to be part of this work, which won the best paper award in #SIGGRAPHAsia this year!

@MinhyukSung Thank you for visiting us!!

RT @zianwang97: 🚀 Introducing our #SIGGRAPHAsia work “Adaptive Shells”, a novel #NeRF formulation that yields high visual fidelity and grea…

RT @haosu_twitr: 📢Thrilled to announce sudoAI (@sudoAI_), founded by a group of leading AI talents and me!🚀 We are dedicated to revolution…

Looking through the results, they are amazing!!

FlexiCubes is covered by @twominutepapers! It's an advanced version of DMTet, can produce uniform mesh tessellation, and align with sharp edges (led by @TianchangS). Code integration with GET3D & Nvdiffrec