Ruowen Zhao

About

Ruowen Zhao is a rising researcher in 3D computer vision and autonomous driving perception, whose work centers on implicit neural representations for scene understanding from sparse sensor data. Her most notable contribution is the development of LODE (Locally Conditioned Eikonal Implicit Scene Completion), a groundbreaking method that reconstructs dense 3D scenes from incomplete LiDAR point clouds. By introducing a locally conditioned eikonal formulation, Zhao’s approach enables robots to detect multi-scale obstacles and analyze object occlusions with unprecedented accuracy—a critical capability for safe autonomous navigation. This work, published in 2023, has already garnered 26 citations, reflecting its immediate impact on the field. Zhao’s research bridges the gap between sparse real-world sensor inputs and the dense, continuous scene representations needed for robust perception. Her achievements highlight a talent for tackling fundamental challenges in implicit representation learning, positioning her as a promising voice in the next generation of autonomous systems researchers.

Research Focus

Key Achievements

Top Papers

1. 1
   LODE: Locally Conditioned Eikonal Implicit Scene Completion from Sparse LiDAR

   26 citations · 2023

Key Collaborators

Contact & Links