Renjie Liao

Papers

Total Citations

211

H-Index

About

Renjie Liao is a researcher working at the intersection of deep learning, graph-based representations, and autonomous systems. His most recognized contribution is **LaneRCNN**, a framework for graph-centric motion forecasting that addresses one of the most challenging problems in self-driving technology: predicting the future behaviors of dynamic actors such as vehicles and pedestrians. By leveraging distributed graph representations, LaneRCNN captures the complex interactions between actors, their intentions, and surrounding map structures, offering a principled approach to trajectory prediction in real-world driving scenarios. Liao's work has garnered significant attention in the robotics and autonomous driving communities, with his motion forecasting research accumulating over 200 citations, reflecting its practical relevance and methodological novelty. His research demonstrates a strong command of geometric deep learning and structured prediction, areas that are increasingly critical as autonomous systems must reason about dynamic, spatially rich environments. For students and researchers entering fields such as self-driving cars, robot navigation, or graph neural networks, Liao's work offers foundational insights into how structured, graph-based models can be effectively applied to real-time, safety-critical prediction tasks.