Yushi Lan

I am currently a 4th-year Ph.D. candidate at MMLab@NTU, Nanyang Technological University, supervised by Prof. Chen Change Loy and working closely with Dr. Daibo. I got my bachelor degree in software engineering from Yepeida Honors College, Beijing Univ of Posts and Tele (BUPT) in 2020.

Email  /  CV  /  Github

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.

SAR3D generates and understands 3D object in a scale-level autoregressive way.

ObjCtrl-2.5D enables training-free, precise, and versatile object control in I2V generation by using 3D depth and camera trajectories.

GaussianAnything generates high-quality and editable surfel Gaussians through a cascaded 3D diffusion pipeline, given single-view images or texts as the conditions.

MvDrag3D provide a precise, generative, and flexible solution for 3D drag-based editing.

3DTopia-XL scales high-quality 3D asset generation using Diffusion Transformer (DiT) built upon an expressive and efficient 3D representation, PrimX.

LN3Diff is a native 3D diffusion model that creates high-quality 3D object mesh from image or text within 8 seconds.

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.

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.

We propose DeformToon3D, an 3D toonification methods that achieves high-quality geometry and texture stylization under given styles.

We propose E3DGE, an encoder-based 3D GAN inversion framework that yields high-quality shape and texture reconstruction.

EVA3D is a high-quality unconditional 3D human generative model that only requires 2D image collections for training.

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.

We propose MagnifierNet, a triple-branch network which accurately mines details from whole to parts in person re-identification (ReID).

Design and source code from Jon Barron's website