Representation and Optimal Recognition of Human Activities

Somboon Hongeng


Abstract

Towards the goal of realizing a generic automatic human activity recognition system, a new formalism is proposed. Activities are described by a chained hierarchical representation using three type of entities: image features, mobile object properties and scenarios. Taking image features of tracked moving regions from an image sequence as input, mobile object properties are first computed by specific methods while noise is suppressed by statistical methods. Scenarios are recognized from mobile object properties based on Bayesian analysis. A sequential occurrence several scenarios are recognized by an algorithm using a probabilistic finite-state automaton (a variant of structured HMM). Finally, the validity and the effectiveness of our approach is demonstrated on both real-world and perturbed data.

On-line references

Video Surveillance And Monitoring


Maintained by Alexandre R.J. FRANÇOIS