Defu Cao
Papers
3
Total Citations
69
H-Index
3
About
Defu Cao is a researcher at the forefront of machine learning for spatiotemporal and graph-structured data, with a primary focus on advancing autonomous systems. His most impactful work centers on trajectory prediction—a critical challenge for autonomous vehicles and social mobile robots. Cao’s key contribution is the development of the Spectral Temporal Graph Neural Network (STGNN), a novel framework that models the complex, dynamic interactions between agents in a scene by combining spectral graph theory with temporal reasoning. This approach enables more accurate and context-aware motion forecasting, addressing the difficulty of predicting behavior influenced by both an agent’s own intentions and its surrounding environment. His foundational 2021 paper on this topic has garnered 58 citations, underscoring its influence in the field. By effectively integrating spatial and temporal dependencies, Cao’s work provides a powerful tool for safer and more reliable autonomous navigation, marking him as a rising contributor to the intersection of graph neural networks and intelligent transportation systems.
Research Focus
Key Achievements
Top Papers
- 1Spectral Temporal Graph Neural Network for Trajectory Prediction58 citations · 2021
- 2Spectral Temporal Graph Neural Network for Trajectory Prediction7 citations · 2024
- 3Spectral Temporal Graph Neural Network for Trajectory Prediction4 citations · 2021