Research works

Image Presentation

  • PDBN Based Human Action Recognition
  • Fusion of Multimodal Information for Event Detection
  • Test Bed: Seminar Environment of On-the-Spot Archiving

Dynamic Bayesian Network is a very good tool to fuse many kinds of information for natural action recognition. During the recognition of dynamic system, different structures of Dynamic Bayesian Network are designed for different problems, in which many kinds of weak information are fused to be strong information. And according to the source, confidence and importance, each kind of information is given a weight, which is very useful for optimization and inference with multi-information. We propose to combine primitive features and Dynamic Bayesian Networks to improve the recognition accuracy.

Introduction about this research topic is to be added.

Introduction about this research topic is to be added.