Research
My interests lie in the intersection of computer vision, computer graphics, and machine learning,
particularly in inverse graphics powered by neural rendering,
including 3D generative models, shape analysis and 3D avatar, etc.
Your browser does not support the video tag.
GaussianAnything: Interactive Point Cloud Latent Diffusion for 3D Generation
Yushi Lan ,
Shangchen Zhou ,
Zhaoyang Lyu ,
Fangzhou Hong ,
Shuai Yang ,
Bo Dai ,
Xingang Pan ,
Chen Change Loy
preprint , 2024
project page
/
arXiv
/
Code
GaussianAnything generates high-quality and editable surfel Gaussians through a cascaded 3D diffusion pipeline, given single-view images or texts as the conditions.
Your browser does not support the video tag.
MVDrag3D: Drag-based Creative 3D Editing via Multi-view Generation-Reconstruction Priors
Honghua Chen ,
Yushi Lan ,
Yongwei Chen ,
Yifan Zhou ,
Xingang Pan ,
preprint , 2024
project page
/
arXiv
/
Code
MvDrag3D provide a precise, generative, and flexible solution for 3D drag-based editing.
3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion
Zhaoxi Chen ,
Jiaxiang Tang ,
Yuhao Dong ,
Ziang Cao ,
Fangzhou Hong ,
Yushi Lan ,
Tengfei Wang ,
Haozhe Xie ,
Tong Wu ,
Shunsuke Saito ,
Liang Pan ,
Dahua Lin ,
Ziwei Liu ,
preprint , 2024
project page
/
arXiv
/
Code
3DTopia-XL scales high-quality 3D asset generation using Diffusion Transformer (DiT) built upon an expressive and efficient 3D representation, PrimX .
Your browser does not support the video tag.
LN3Diff: Scalable Latent Neural Fields Diffusion for Speedy 3D Generation
Yushi Lan ,
Fangzhou Hong ,
Shuai Yang ,
Shangchen Zhou ,
Xuyi Meng ,
Bo Dai ,
Xingang Pan ,
Chen Change Loy
ECCV , 2024
project page
/
arXiv
/
Code
LN3Diff is a native 3D diffusion model that creates high-quality 3D object mesh from image or text within 8 seconds.
Gaussian3Diff: 3D Gaussian Diffusion for 3D Full Head Synthesis and Editing
Yushi Lan ,
Feitong
Tan ,
Di Qiu ,
Qiangeng Xu
Kyle Genova
Zeng Huang ,
Sean Fanello ,
Rohit Pandey ,
Thomas Funkhouser ,
Chen Change Loy ,
Yinda Zhang
ECCV , 2024
project page
/
arXiv
Gaussian3Diff adopts 3D Gaussians defined in UV space as the underlying 3D
representation, which intrinsically support high-quality novel view synthesis, 3DMM-based animation
and 3D diffusion for unconditional generation.
Learning Dense Correspondence for NeRF-Based Face Reenactment
Songlin Yang ,
Wei Wang ,
Yushi Lan ,
Xiangyu Fan ,
Bo Peng ,
Lei Yang ,
Jing Dong
AAAI , 2024
project page /
arXiv
We propose a novel face reenactment framework,
which adopts tri-planes as fundamental NeRF representation and decomposes face tri-planes into three
components: canonical tri-planes, identity deformations, and motion.
Your browser does not support the video tag.
DeformToon3D: Deformable 3D Toonification from Neural Radiance Fields
Junzhe Zhang* ,
Yushi Lan* ,
Shuai Yang ,
Fangzhou Hong ,
Quan Wang ,
Chai Kiat Yeo ,
Ziwei Liu ,
Chen Change Loy
ICCV , 2023
project page
/
arXiv
/
Code
We propose DeformToon3D, an 3D toonification methods that achieves high-quality geometry and texture
stylization under given styles.
E3DGE: Self-Supervised Geometry-Aware Encoder for Style-Based 3D GAN Inversion
Yushi Lan ,
Xuyi Meng ,
Shuai Yang ,
Chen Change Loy ,
Bo Dai
CVPR , 2023
project page
/
arXiv
/
video
/
Code
We propose E3DGE, an encoder-based 3D GAN inversion framework that yields high-quality shape and
texture reconstruction.
Your browser does not support the video tag.
EVA3D: Compositional 3D Human Generation from 2D Image Collections
Fangzhou Hong ,
Zhaoxi Chen ,
Yushi Lan ,
Liang
Pan ,
Ziwei Liu
ICLR , 2023, Spotlight
project page
/
arXiv
/
video
/
Code
EVA3D is a high-quality unconditional 3D human generative model that only requires
2D image collections for training.
Your browser does not support the video tag.
DDF: Correspondence Distillation from NeRF-Based GAN
Yushi Lan ,
Chen Change Loy ,
Bo Dai
IJCV , 2022
project page
/
arXiv
/
Springer
We study dense correspondence, which plays a key role in 3D scene understanding but has been ignored in NeRF
research.
DDF presents a novel way to distill dense NeRF correspondence from pre-trained NeRF GAN
unsupervisedly.
Your browser does not support the video tag.
Magnifier: Towards Semantic Adversary and Fusion for Person Re-identification
Yushi Lan* ,
Yuan Liu* ,
Xinchi Zhou ,
Maoqing Tian ,
Xuesen Zhang ,
Shuai Yi ,
Hongsheng Li ,
BMVC , 2020
arXiv
/
Code
We propose MagnifierNet, a triple-branch network which accurately mines details from whole to parts in
person re-identification (ReID).