The central goal of my research is to build intelligent machines that are capable of recreating our visual world. I strive to achieve this by establishing a connection between the visual contents from physical world and from virtual scenes. My research helps people better restore our visual world in the digital format and render artistic effects more easily. Especially, I study three main topics:

Generative Model: Generative Adversarial Networks, Diffusion Model, Variational Autoencoder
Neural Rendering: Differentiable Rendering, Image-based Rendering, Neural Radiance Fields
Computational Phtography: Novel-view Synthesis, Inverse Problem, 2D/3D Editing and Manipulation