Ruochen Li's Outputs (4)

BP-SGCN: Behavioral Pseudo-Label Informed Sparse Graph Convolution Network for Pedestrian and Heterogeneous Trajectory Prediction (2025)
Journal Article
Li, R., Katsigiannis, S., Kim, T.-K., & Shum, H. P. H. (online). BP-SGCN: Behavioral Pseudo-Label Informed Sparse Graph Convolution Network for Pedestrian and Heterogeneous Trajectory Prediction. IEEE Transactions on Neural Networks and Learning Systems, https://doi.org/10.1109/TNNLS.2025.3545268

Trajectory prediction allows better decision-making in applications of autonomous vehicles (AVs) or surveillance by predicting the short-term future movement of traffic agents. It is classified into pedestrian or heterogeneous trajectory prediction.... Read More about BP-SGCN: Behavioral Pseudo-Label Informed Sparse Graph Convolution Network for Pedestrian and Heterogeneous Trajectory Prediction.

Unified Spatial-Temporal Edge-Enhanced Graph Networks for Pedestrian Trajectory Prediction (2025)
Journal Article
Li, R., Qiao, T., Katsigiannis, S., Zhu, Z., & Shum, H. P. (online). Unified Spatial-Temporal Edge-Enhanced Graph Networks for Pedestrian Trajectory Prediction. IEEE Transactions on Circuits and Systems for Video Technology, https://doi.org/10.1109/TCSVT.2025.3539522

Pedestrian trajectory prediction aims to forecast future movements based on historical paths. Spatial-temporal (ST) methods often separately model spatial interactions among pedestrians and temporal dependencies of individuals. They overlook the dire... Read More about Unified Spatial-Temporal Edge-Enhanced Graph Networks for Pedestrian Trajectory Prediction.

From Category to Scenery: An End-to-End Framework for Multi-Person Human-Object Interaction Recognition in Videos (2024)
Presentation / Conference Contribution
Qiao, T., Li, R., Li, F. W. B., & Shum, H. P. H. (2024, December). From Category to Scenery: An End-to-End Framework for Multi-Person Human-Object Interaction Recognition in Videos. Presented at ICPR 2024: International Conference on Pattern Recognition, Kolkata, India

Video-based Human-Object Interaction (HOI) recognition explores the intricate dynamics between humans and objects, which are essential for a comprehensive understanding of human behavior and intentions. While previous work has made significant stride... Read More about From Category to Scenery: An End-to-End Framework for Multi-Person Human-Object Interaction Recognition in Videos.

Multiclass-SGCN: Sparse Graph-based Trajectory Prediction with Agent Class Embedding (2022)
Presentation / Conference Contribution
Li, R., Katsigiannis, S., & Shum, H. P. (2022, October). Multiclass-SGCN: Sparse Graph-based Trajectory Prediction with Agent Class Embedding. Presented at ICIP 2022: IEEE International Conference in Image Processing, Bordeaux, France

Trajectory prediction of road users in real-world scenarios is challenging because their movement patterns are stochastic and complex. Previous pedestrian-oriented works have been successful in modelling the complex interactions among pedestrians, bu... Read More about Multiclass-SGCN: Sparse Graph-based Trajectory Prediction with Agent Class Embedding.