A computer vision integration model for a multi-modal cognitive system