Method and apparatus for detecting suspicious activity using video analysis from alternative camera viewpoint

Abstract

A system performs video identification of items involved with transactions, as captured in video data, and compares this item identification information with transaction data obtained from a transaction terminal such as a point-of-sale register to identify situations that are suspicious and may indicate fraudulent activity or operator error. In particular, the system as disclosed herein utilizes unique camera positions, and can use video data from more than one camera, to perform robust detections of fraudulent or suspicious activities associated with transactions in the transaction area. By obtaining video data originating from at least one video camera that monitors a transaction area, the system can automatically (i.e. no human involvement needed) analyze the video data to track items involved in a transaction in the transaction area. Using this information, the system can compare the video analysis of the tracked items to transaction data produced from a transaction terminal to identify suspicious activity. The system can also create, update and maintain object models for items tracked in a transaction area, in part, to assist in identifying suspicious activity.