Lip shape extraction for word recognition by using hardware active contour model

Even in noisy environments such as in machine factories or in crushes, we are supposed to control equipment by using human voices. Although word recognition techniques have been successfully developed, noises in environments causes serious recognition problems. In such cases, lip shape movements are expected to be useful as supporting data to improve the performance of recognition. In this paper, a method to extract outer and inner lip shape movements from input face images by using an active contour model is discussed. The proposed method is implemented as hardware circuits in an FPGA chip to process NTSC video signals in real-time. Experimental results of lip shape extraction are shown to confirm the effectiveness of the proposed method.