A MODULAR NEURAL-NETWORK VECTOR PREDICTOR FOR PREDICTIVE IMAGE-CODING

In this paper, we present a modular neural network vector predictor th at improves the predictive component of a predictive vector quantizati on (PVQ) scheme. The proposed vector prediction technique consists of five dedicated predictors (experts), where each expert predictor is op timized for a particular class of input vectors. An input vector is cl assified into one of five classes, based on its directional variances. One expert predictor is optimized for stationary blocks, and each of the other four expert predictors are optimized to predict horizontal, vertical, 45 degrees, and 135 degrees diagonally oriented edge-blocks, respectively. An integrating unit is then used to select or combine t he outputs of the experts in order to form the final output Of the mod ular network, Therefore, no side information is transmitted to the rec eiver about the selected predictor or the integration of the predictor s. Experimental results show that the proposed scheme gives an improve ment of 1.7 dB over a single multilayer perceptron on (MLP) predictor, Furthermore, if the information about the predictor selection is sent to the receiver, the improvement could bit up to 3 dB over a single M LP predictor. The perceptual quality of the predicted images is also s ignificantly improved.