Automatic learning of an activity-based semantic scene model
- 24 January 2004
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
The paper proposes an activity-based semantic model for a scene under visual surveillance. It illustrates methods that allow unsupervised learning of the model from trajectory data derived from automatic visual surveillance cameras. Results are shown for each method. Finally, the benefits of such a model in a visual surveillance system are discussed.Keywords
This publication has 7 references indexed in Scilit:
- Coupled hidden Markov models for complex action recognitionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Path detection in video surveillanceImage and Vision Computing, 2002
- Spatial and Probabilistic Modelling of Pedestrian BehaviourPublished by British Machine Vision Association and Society for Pattern Recognition ,2002
- Learning Variable-Length Markov Models of BehaviorComputer Vision and Image Understanding, 2001
- TRACKING AND SURVEILLANCE IN WIDE-AREA SPATIAL ENVIRONMENTS USING THE ABSTRACT HIDDEN MARKOV MODELInternational Journal of Pattern Recognition and Artificial Intelligence, 2001
- Finding Paths in Video SequencesPublished by British Machine Vision Association and Society for Pattern Recognition ,2001
- Generation of semantic regions from image sequencesPublished by Springer Nature ,1996