Learn More
In this paper we introduce a real-time system for action detection. The system uses a small set of robust features extracted from 3D skeleton data. Features are effectively described based on the probability distribution of skeleton data. The descriptor computes a pyramid of sample covariance matrices and mean vectors to encode the relationship between the(More)
Cheap and readily available storage systems help us gather hundreds of terabytes of data everyday. This information can be analyzed to understand and predict radical human behavior and actions in order to mitigate loss of lives and property. We propose using machine learning method on CCTV, audio, and geotagged Twitter data to identify suspicious persons(More)
  • 1