We propose a new method for distinguishing gazing direction of human eyes using neural network (NN).Firstly,ten primary parameters are extracted from human face,then after some simple computation we get feature parameters which can present gazing direction.NN is proposed to recognize gazing direction using these parameters.Experimental results and analyses are presented to demonstrate the efficacy of the algorithm.Because of the simple computation and the small scale NN,the system can distinguish gazing direction in real time at high speed,and the recognition rate can be high.