AIMI Research Meeting: Neural Rendering: A Round Trip from Pixel to Model - Chuan Li, PhD

via Zoom - email aimicenter@stanford.edu for link

Stanford community and AIMI Affiliates only

Abstract:
In this talk, we will discuss modern rendering techniques and discuss how deep learning can extend the gamut of this long-lasting research topic. We will investigate deep neural networks for both forward and inverse graphics. In particular, we will focus on "differentiable rendering," a methodology that solves complex inverse graphics problems and achieved great success in applications such as image restoration, generation, and editing.

About:
Chuan Li has over 10 years of research and engineering experience in the convergent fields of computer vision, computer graphics, and machine learning, specifically on the subject of visual data analysis and synthesis. He received Ph.D. in computer science from the University of Bath. He worked at Max Planck Institute of Informatics and Utrecht University as Postdoc. His work appeared in top research conferences and journals (NIPS, ICCV, CVPR, Siggraph, etc.) and media press such as BBC, Times magazine, and New Scientists. Currently, he works at Lambda labs. Inc, a San Francisco-based AI company that provides infrastructure and consultancy service for machine learning.