
PhD at UCSD @HaoSuLabUCSD Previous intern @AdobeResearch @Hillbot @NVIDIA
San Diego, CA
Joined July 2017
- Tweets21
- Following146
- Followers266
- Likes19
Pinned Tweet
🚀 Our new convex decomposition CuACD! (SIGGRAPH Asia 2026)
⚡ Fully GPU-resident
⚡ Matches CoACD quality with ~100x speed-up
⚡ pip install cuacd
✅ While CoACD finishes one mesh, CuACD finishes a dozen
Paper: arxiv.org/abs/2609.28731
Code: github.com/eliphatfs/cuacd
Xinyue Wei retweeted
Our robot is live at ICRA! Come visit Booth 171 to see our zero-shot pick-and-place demo in action. Bring an item and test it yourself.
🚀 Our new GPU-based mesh optimization -- PaMO!
⚡ Parallel remeshing → handle arbitrary meshes
⚡ Robust intersection-free simplification
⚡ Safe projection → sharp features restored
✅ 2M → 2k triangles in 2.29s on RTX 4090
Check the code here: github.com/SarahWeiii/pamo
1
7
270
Really honored to have contributed to the physics alignment evaluation of the Cosmos during my internship at NVIDIA 💚
The model is now officially released — feel free to check it out!
Whether you're a researcher or developer, #NVIDIACosmos world foundation models are now openly available under our permissive license to the physical AI community via NGC & @huggingface. 🤗 #CES2025
See how Cosmos is democratizing #physicalAI development: nvda.ws/3DJwE5v
1
24
2,926
Fantastic work! Thrilled to see our method CoACD being cited and contributing to advancements in geometry processing. It's always exciting to see how our research can support the community. 🚀✨
Everything you love about generative models — now powered by real physics!
Announcing the Genesis project — after a 24-month large-scale research collaboration involving over 20 research labs — a generative physics engine able to generate 4D dynamical worlds powered by a physics simulation platform designed for general-purpose robotics and physical AI applications.
Genesis's physics engine is developed in pure Python, while being 10-80x faster than existing GPU-accelerated stacks like Isaac Gym and MJX. It delivers a simulation speed ~430,000 faster than in real-time, and takes only 26 seconds to train a robotic locomotion policy transferrable to the real world on a single RTX4090 (see tutorial: genesis-world.readthedocs.io…).
The Genesis physics engine and simulation platform is fully open source at github.com/Genesis-Embodied-…. We'll gradually roll out access to our generative framework in the near future.
Genesis implements a unified simulation framework all from scratch, integrating a wide spectrum of state-of-the-art physics solvers, allowing simulation of the whole physical world in a virtual realm with the highest realism.
We aim to build a universal data engine that leverages an upper-level generative framework to autonomously create physical worlds, together with various modes of data, including environments, camera motions, robotic task proposals, reward functions, robot policies, character motions, fully interactive 3D scenes, open-world articulated assets, and more, aiming towards fully automated data generation for robotics, physical AI and other applications.
Open Source Code: github.com/Genesis-Embodied-…
Project webpage: genesis-embodied-ai.github.i…
Documentation: genesis-world.readthedocs.io…
1/n
Xinyue Wei retweeted
Meet #HillbotAlpha, the first fully autonomous mobile manipulation robot trained using sim-to-real technology.
Designed in Hillbot’s San Diego headquarters, Hillbot Alpha represents the potential of data synthesis via simulation in robotics. Through a robust strategy that combines a small sample of real-world data with synthetic data, Hillbot Alpha can effectively adapt to evolving task requirements and environments while working safely beside humans.
#Hillbot #AGI #EmbodiedAI #AI #Sim2Real #ArtificialIntelligence #Simulation #Simtoreal
Xinyue Wei retweeted
Want high-quality 3D meshes with sharp geometric details? Try our newly released MeshFormer!
It only takes 8 GPUs for two days of training, outperforming state-of-the-art models that use over a hundred GPUs!
With 3D-native input guidance, representations, supervision, and post-processing, we significantly improve the training efficiency and geometric quality of feed-forward reconstruction models!
Project page: meshformer3d.github.io
@iam_NCJ
🚀 Thrilled to announce the release of the reproduced MeshLRM demo! 🎉 Generate textured 3D meshes from one or more unposed images in seconds. Check it out: huggingface.co/spaces/sudo-a…
Xinyue Wei retweeted
Want to obtain time-consistent dynamic mesh from monocular videos? Introducing:
Dynamic Gaussians Mesh: Consistent Mesh Reconstruction from Monocular Videos
liuisabella.com/DG-Mesh/
We reconstruct meshes with flexible topology change and build the corresp. across meshes. 🧵(1/n)
Check our MeshLRM 🌟, which achieves state-of-the-art mesh reconstruction from sparse-view inputs within 1 second! 🚀🚀🚀
Paper: arxiv.org/abs/2404.12385
Project page: sarahweiii.github.io/meshlrm…
🥳 ZeroRF has been accepted to #CVPR2024!
Check the code here: github.com/eliphatfs/zerorf
Introducing ZeroRF, where we reconstruct high-quality radiance fields from *sparse views* at an order of magnitude faster than previous methods (30 secs for 320x320) without any pretraining or additional regularization!
sarahweiii.github.io/zerorf/
Xinyue Wei retweeted
Introducing ZeroRF, where we reconstruct high-quality radiance fields from *sparse views* at an order of magnitude faster than previous methods (30 secs for 320x320) without any pretraining or additional regularization!
sarahweiii.github.io/zerorf/
Announcing DISO - a fast, efficient, and differentiable iso-surface extraction package🤩!
DISO outperforms DMTet & FlexiCubes by 10-20x in speed and demands significantly less CUDA memory. It is also easily installable from pip.
Check it out!
github.com/SarahWeiii/diso
Xinyue Wei retweeted
One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion
paper page: huggingface.co/papers/2311.0…
Recent advancements in open-world 3D object generation have been remarkable, with image-to-3D methods offering superior fine-grained control over their text-to-3D counterparts. However, most existing models fall short in simultaneously providing rapid generation speeds and high fidelity to input images - two features essential for practical applications. In this paper, we present One-2-3-45++, an innovative method that transforms a single image into a detailed 3D textured mesh in approximately one minute. Our approach aims to fully harness the extensive knowledge embedded in 2D diffusion models and priors from valuable yet limited 3D data. This is achieved by initially finetuning a 2D diffusion model for consistent multi-view image generation, followed by elevating these images to 3D with the aid of multi-view conditioned 3D native diffusion models. Extensive experimental evaluations demonstrate that our method can produce high-quality, diverse 3D assets that closely mirror the original input image.
Xinyue Wei retweeted
📢Thrilled to announce sudoAI (@sudoAI_), founded by a group of leading AI talents and me!🚀
We are dedicated to revolutionizing digital & physical realms by crafting interactive AI-generated 3D environments!
Join our 3D Gen AI model waitlist today!
👉sudo.ai/
Xinyue Wei retweeted
Excited to release Zero123++, an image-conditioned diffusion model for generating high-quality, 3D-consistent multi-view images from a single input view!
Paper: arxiv.org/abs/2310.15110
Code: github.com/SUDO-AI-3D/zero12…
🤗Demo: huggingface.co/spaces/sudo-a…
Xinyue Wei retweeted
The ManiSkill 2022 large-scale robotic manipulation challenge is officially released🚀!
Feature:
- Rigid & soft body tasks
- Complex contact-rich dynamics
- Diverse object categories & geometry
- Fast simulation for visual RL⚡️(5x faster than DMControl)
sapien.ucsd.edu/challenges/m…
Xinyue Wei retweeted
How to get started in embodied AI research? How to build a virtual environment for your next project in navigation, manipulation, locomotion, or rearrangement?
Welcome to our #CVPR2022 tutorial Building and Working in Environments for Embodied AI!
ai-workshops.github.io/build…
🥳Very excited to share our new work: CoACD, a novel convex decomposition method. It better preserves collision conditions of the input shape with fewer components, enabling delicate and efficient object interaction in physics simulators! #SIGGRAPH2022
colin97.github.io/CoACD/













