Kalman Filtering

Kalman Filtering is one of the well-established methods in objects tracking. The basic jargons used in this method are:
1) state variables (these represent the current properties of the interested object)
2) measurement (this refers to actual measurements obtained from objects detection)
3) white noise (this refers to unwanted information obtained from objects detection (eg. noises due to illumintation changes, errors due to unaccurated detection, etc).

Good Link: http://www.innovatia.com/software/papers/kalman.htm