A Comparison of Vector Quantization Codebook Generation Algorithms Applied to Automatic Face Recognition

Automatic facial recognition is an attractive solution to the problem of computerised personal identification. In order to facilitate a cost effective solution, high levels of data reduction are required when storing the facial information. Vector Quantization has previously been used as a data reduction technique for the encoding of facial images. This paper identifies the fundamental importance of the vector quantizer codebooks in the performance of the system. Two different algorithms the Linde-Buzo-Gray algorithm and Kohonen's Self Organising Feature Map have been used to obtain two sets of facial feature codebooks. For comparison, the system performance has also been analysed using a codebook dedicated to the test population. It has been shown that by using a good codebook generation algorithm it is possible to substantially reduce the dimensionality of the vector codebooks, with remarkably little degradation in system performance.