My name is Zixuan Huang (黄子煊, pronounced as Zi-Shwan Hwung), currently a fourth-year CS PhD student at University of Illinois Urbana-Champaign advised by James Rehg. I got my masters degree at University of Wisconsin-Madison, where I was fortunate to work with Professor Yin Li. I received my B.E. in the Special Class for Gifted Young from University of Science and Technology of China (USTC) in 2018. My main research focus is scalable and generalizable 3D shape reconstruction.
In the past few years, I have also worked with Varun Jampani and Mark Boss at Stability AI, Chao-Yuan Wu and Justin Johnson at FAIR Labs (Meta AI), Varun Jampani and Yuanzhen Li at Google Research, as well as Hongsheng Li and Shuai Yi at Sensetime Research.
SF3D: Stable Fast 3D Mesh Reconstruction with UV-unwrapping and Illumination Disentanglement
SF3D takes a single image as input and generates a textured UV-unwrapped 3D model in under a second.
Mark Boss, Zixuan Huang, Aaryaman Vasishta, Varun Jampani
Arxiv Preprint, 2024
paper / project page / code / model / demo
TripoSR: Fast 3D Object Reconstruction from a Single Image
Large open-source model for high-quality 3D reconstruction from a single image under one second.
Dmitry Tochilkin, David Pankratz, Zexiang Liu, Zixuan Huang, Adam Letts, Yangguang Li, Ding Liang, Christian Laforte, Varun Jampani, Yan-Pei Cao
Tech Report, 2024
PointInfinity: Resolution-Invariant Point Diffusion Models
Point diffusion model that trains on low resolution point clouds, while generates faithful high resolution point clouds. Performance continuously improves as inference resolution increases.
Zixuan Huang, Justin Johnson, Shoubhik Debnath, James M. Rehg, Chao-Yuan Wu
CVPR 2024
ZeroShape: Regression-based Zero-shot Shape Reconstruction
SOTA 3D shape reconstructor with high computational efficiency and low training data budget.
Zixuan Huang*, Stefan Stojanov*, Anh Thai, Varun Jampani, James M. Rehg
CVPR 2024
paper / code / project page / demo
ShapeClipper: Scalable 3D Shape Learning from Single-View Images via Geometric and CLIP-based Consistency
CLIP and geometric consistency constraints facilitate scalable learning of object shape reconstruction.
Zixuan Huang, Varun Jampani, Anh Thai, Yuanzhen Li, Stefan Stojanov, James M. Rehg
CVPR 2023
paper / code / project page / video
Low-shot Object Learning with Mutual Exclusivity Bias
Mutual Exclusivity Bias enables fast learning of objects that generalizes.
Anh Thai, Ahmad Humayun, Stefan Stojanov, Zixuan Huang, Bikram Boote, James M. Rehg
NeurIPS 2023, Datasets and Benchmarks Track
paper / code / project page
Learning Dense Object Descriptors from Multiple Views for Low-shot Category Generalization
Multi-view self-supervised learning that allows for low-shot category recognition.
Stefan Stojanov, Anh Thai, Zixuan Huang, James M. Rehg
NeurIPS 2022
paper / code / project page / poster / video
Planes vs. Chairs: Category-guided 3D shape learning without any 3D cues
A 3D-unsupervised model that learns shapes of multiple object categories at once.
Zixuan Huang, Stefan Stojanov, Anh Thai, Varun Jampani, James M. Rehg
ECCV 2022
paper / code / project page / poster / video
The Surprising Positive Knowledge Transfer in Continual 3D Object Shape Reconstruction
Continual learning of 3D shape reconstruction does not suffer from catastrophic
forgetting as much as discriminative learning tasks.
Anh Thai, Stefan Stojanov, Zixuan Huang, James M. Rehg
3DV 2022
HMS-Net: Hierarchical Multi-scale Sparsity-invariant Network for Sparse Depth Completion
A Multi-scale sparsity-invariant network for monocular depth completion.
Zixuan Huang, Junming Fan, Shenggan Cheng, Shuai Yi, Xiaogang Wang, Hongsheng Li
IEEE TIP 2019