Towards an online detection of pedestrian flocks in urban canyons by smoothed spatio-temporal clustering of GPS trajectories.
Martin WirzPablo SchläpferMikkel Baun KjærgaardDaniel RoggenSebastian FeeseGerhard TrösterPublished in: GIS-LBSN (2011)
Keyphrases
- spatio temporal
- object detection
- clustering algorithm
- moving objects
- motion patterns
- detection algorithm
- spatial and temporal
- crowded scenes
- pedestrian detection
- trajectories of moving objects
- real time
- urban areas
- hierarchical clustering
- automatic detection
- temporal segmentation
- space time
- unsupervised learning
- online learning
- k means
- clustering method
- anomaly detection
- image sequences
- detection method
- false positives
- object trajectories
- detection rate
- motion trajectories
- spatio temporal data
- view invariant
- moving objects databases
- data clustering
- event detection
- self organizing maps
- data points
- computer vision
- human detection
- man made
- change detection
- outlier detection
- spatio temporal databases
- video sequences