Assist. Prof. Dr . Shengjie Li | Internet of Things | Editorial Board Member

Beijing University of Posts and Telecommunications | China

Assist. Prof. Dr. Li Shengjie is a researcher specializing in robust visual object tracking, focusing on resolving key challenges such as occlusion, noise, scale variation, drift, and real-time computational constraints in dynamic visual environments. His body of work integrates particle filter methodologies, siamese network models, and correlation filter (CF)-based approaches to develop adaptive and resilient tracking frameworks. He has introduced several influential models, including a part-aware tracking framework that leverages local target structures, a dynamic particle filter architecture for improved failure correction, a noise-aware similarity measurement framework, and an efficient particle scale-space strategy designed to more accurately estimate target size under challenging scale variations. Assist. Prof. Dr. Li’s research also advances lightweight and accelerated particle-filter solutions that significantly reduce computational cost while preserving tracking accuracy, supporting real-time implementation. Additionally, his work integrates ensemble deep features and decision fusion mechanisms to combat model drift and enhance robustness during sudden appearance changes or out-of-view scenarios. His earlier contributions include physical modeling studies, reflecting a progression toward highly refined hybrid deep-learning and filter-based tracking systems. With eight publications, more than 100 citations, and nearly 300 reads, Assist. Prof. Dr. Li Shengjie’s research is recognized for systematically improving the reliability and efficiency of visual tracking algorithms. His innovations consistently address the limitations of traditional tracking methods by introducing adaptive, context-sensitive, and computationally optimized approaches that advance the field toward more effective real-world application.

Profile: Research Gate

Featured Publications

• Li, S., Zhao, S., Cheng, B., & Chen, J. (2023). Part-aware framework for robust object tracking.

• Li, S., Zhao, S., Cheng, B., & Chen, J. (2021). Dynamic particle filter framework for robust object tracking.

• Li, S., Zhao, S., Cheng, B., & Chen, J. (2020). Noise-aware framework for robust visual tracking.

• Li, S., Zhao, S., Cheng, B., & Chen, J. (2020). Efficient particle scale space for robust tracking.

• Li, S., Zhao, S., Cheng, B., & Chen, J. (2018). Robust visual tracking via hierarchical particle filter and ensemble deep features.