Towards Learning a Driving Simulator from the Real World

Speaker

Yiyi Liao is an assistant professor at Zhejiang University. Prior to that, she received her Ph.D. degree from Zhejiang University and subsequently worked as a Postdoc in MPI for Intelligent Systems and the University of Tübingen. Her research interest lies in 3D computer vision, including 3D scene understanding, 3D reconstruction, and 3D-aware image synthesis.

Abstract

A high-fidelity simulation platform has significant value for the development of autonomous driving. Conventional simulation platforms, manually crafted, often deviate from reality and entail substantial costs. In this talk, I will share our recent efforts towards constructing a high-fidelity driving simulation platform from the real world, powered by neural rendering technology. Specifically, I will present our work in both reconstructing and generating urban scenes with photorealistic appearance, rich semantics, and versatile control capabilities.

Video