Recognition of group activities using dynamic probabilistic networks
Dynamic Probabilistic Networks (DPNs) are exploited for modelling the temporal relationships among a set of dif-ferent object temporal events in the scene for a coherent and robust scene-level behaviour interpretation. In particu-lar, we develop a Dynamically Multi-Linked Hidden Markov Model (DML-HM...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Text |
| Language: | English |
| Subjects: | |
| Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.643.8247 http://www.eecs.qmul.ac.uk/~txiang/publications/gong-xiang-iccv03-cameraReady.pdf |
| Summary: | Dynamic Probabilistic Networks (DPNs) are exploited for modelling the temporal relationships among a set of dif-ferent object temporal events in the scene for a coherent and robust scene-level behaviour interpretation. In particu-lar, we develop a Dynamically Multi-Linked Hidden Markov Model (DML-HMM) to interpret group activities involving multiple objects captured in an outdoor scene. The model is based on the discovery of salient dynamic interlinks among multiple temporal events using DPNs. Object temporal events are detected and labelled using Gaussian Mixture Models with automatic model order selection. A DML-HMM is built using Schwarz’s Bayesian Information Cri-terion based factorisation resulting in its topology being in-trinsically determined by the underlying causality and tem-poral order among different object events. Our experiments demonstrate that its performance on modelling group activ-ities in a noisy outdoor scene is superior compared to that of a Multi-Observation Hidden Markov Model (MOHMM), a Parallel Hidden Markov Model (PaHMM) and a Coupled |
|---|