Dafeng Wang
Papers
1
Total Citations
40
H-Index
1
About
Dafeng Wang is a leading researcher in computer vision and autonomous systems, with a core focus on pedestrian trajectory prediction—a critical challenge for safe autonomous driving and socially aware robotics. His most influential work, "SEEM: A Sequence Entropy Energy-Based Model for Pedestrian Trajectory All-Then-One Prediction" (2022), introduces a novel energy-based framework that addresses persistent limitations in the field: lack of trajectory diversity, poor accuracy, and instability. By leveraging sequence entropy, Wang’s model generates more realistic and varied future paths, setting a new benchmark for predictive robustness. With 40 citations in just two years, this paper has quickly become a reference point for researchers tackling multi-modal trajectory forecasting. Wang’s contributions are particularly notable for bridging theoretical energy-based models with practical, real-world deployment needs, offering a principled solution to the "all-then-one" prediction paradigm. His work not only advances the technical frontier but also directly impacts the reliability of autonomous navigation systems, making him a rising voice in the intersection of machine learning and embodied AI.
Research Focus
Key Achievements
Top Papers
- 1