Zishan Zhao
Papers
3
Total Citations
55
H-Index
3
About
Zishan Zhao is an emerging researcher specializing in pedestrian trajectory prediction, with a particular focus on developing advanced deep learning architectures that capture the complex spatial and temporal dynamics of human movement. His work sits at the intersection of computer vision, graph neural networks, and autonomous systems, addressing one of the most challenging problems in intelligent transportation and robot navigation. Zhao's most significant contributions include the development of STIGCN (Spatial-Temporal Interaction-aware Graph Convolution Network), which innovatively disentangles social interaction factors from individual pedestrian movement patterns to improve prediction accuracy. His follow-up work, DSTCNN (Deformable Spatial-Temporal Convolutional Neural Network), further advances the field by introducing deformable convolutions to better model irregular and dynamic pedestrian behaviors. Both papers have accumulated 26 citations each, reflecting strong and rapid community recognition for work published within just two years. These contributions are particularly valuable for real-world applications in autonomous driving and robotic navigation, where accurate pedestrian trajectory forecasting is safety-critical. Zhao's research demonstrates a consistent commitment to architecturally innovative solutions that meaningfully advance the state of the art in human motion prediction.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3