Cameras capture time-stamped images of predefined areas (zones) and assets (objects). Behaviors and actions of persons are tracked in the images with respect to the zones and the objects based on rules. Persons are identified when the behaviors or actions indicate that the persons are attempting to access a particular zone or a particular object. Multifactor authentication is performed on the persons based on the rules and access to the particular zone or the particular object is granted or denied. All access attempts along with the corresponding images associated with the access attempts are logged for auditing.