Haifeng Sang
Papers
10
Total Citations
106
H-Index
5
About
Haifeng Sang is a prominent researcher specializing in pedestrian trajectory prediction and intelligent robotics, with a particular focus on graph neural network architectures for modeling complex human movement patterns. His work addresses two of the field's most persistent challenges: accurately capturing social interactions among pedestrians and encoding individual movement factors to produce reliable trajectory forecasts — capabilities critical for autonomous driving, service robotics, and crowd monitoring systems. Sang's most impactful contributions include the development of innovative graph convolution frameworks such as STIGCN, DSTCNN, and IMGCN, each garnering 16–26 citations within just one to two years of publication — a testament to their relevance and adoption by the research community. His IMGCN work stands out for its emphasis on interpretability, while GHGNN and MSDHGNN demonstrate his forward-looking exploration of hypergraph neural networks to model higher-order pedestrian group dynamics. Beyond trajectory prediction, Sang has contributed to robotics engineering, notably designing a wireless in-pipe inspection robot for non-destructive image acquisition, which has attracted 11 citations. His 2025 review of graph neural network-based trajectory prediction methods signals his growing role as a synthesizer and thought leader in this rapidly evolving domain, making his profile essential reading for researchers working at the intersection of computer vision, autonomous systems, and human behavior modeling.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4Design of wireless in-pipe inspection robot for image acquisition11 citations · 2022
- 5
- 6
- 7
- 8
- 9
- 10