Computer Assisted Visual InterActive Recognition: Formal Description and Evaluation

In the proposed computer assisted visual interactive recognition (CAVIAR) methodology, a parameterized geometrical model serves as the human-computer communication channel. The iterative CAVIAR process is modelled as a finite state machine. A flower recognition system is implemented based on the proposed methodology. Evaluation on 30 subjects shows that 1) the accuracy of the CAVIAR system is 90% compared to 50% for the machine alone; 2) its recognition time is 10.7 seconds compared to 26.4 seconds for the human alone; 3) it can be initialized with as few as one training sample per class and still achieve high accuracy; 4) it demonstrates a self-learning ability.